Skip to main content

Full text of "Scotland In Dark Age Britain"

See other formats

published 1996 
Scottish Cultural Press 
PO Box 106, Aberdeen AB9 8ZE 
Tel: 01224 583777 • Fax: 01224 575337 

© Contributors 1996 

All rights reserved. No part of this publication may be reproduced, 
stored in a retrieval system, or transmitted, in any form, or by any means, 
electronic, mechanical, photocopying, recording or otherwise without the 
prior permission of Scottish Cultural Press. 

British Library Cataloguing in Publication Data 

A catalogue record for this book is available from the British Library 

ISBN: 1 898218 61 7 

Price £9.00. 

Designed and typeset by Reprographic Services, 
University of St Andrews 

Printed and bound by BPC-AUP Aberdeen Ltd 

cotland in Dark Aqe Britain 

The Proceedings of a Day Conference 
held on 18 February 1995 

Edited by 
Barbara E. Crawford 

St John's House Papers No. 6 

Scilly Isles • 

Fig. 1. Places in Dark Age Britain mentioned in the text 



Barbara E. Crawford 

page uc 

Ckapter 1 
"Protected Space" in Britain and Ireland 
in the Middle Ages 
Wendy Davies 
page 1 

Ckapter 2 
The Early Christian Carved and Inscribed 
Stones of south-west Britain 
Elisabeth Okasha 

page 21 

Ckapter 3 
From Birsay to Tintagel: A Personal View 
Christopher Morris 

page 37 

Ckapter 4 

Trade in the Dark Age West: A Peripheral Activity? 
Ewan Campbell 
page 79 

Ckapter 5 

Place-Names and the Early Church in Eastern Scotland 
Simon Taylor 

page 93 

Ckapter 6 
Iona, Scotland and the Celi De 
Thomas Clancy 
page 111 

Ckapter 7 
The Emergence of the Regnum Scottorum: 
A Carolingian Hegemony? 
Patrick Wormald 
page 131 

Z Britatn 

Chapter 1 

Fig. 1 p. iv Map of places in Dark Age Britain mentioned in the text. 

Fig. 2.1 p. 22 Cross from Sancreed, Cornwall (Royal Commission on 
Historical Monuments of England). 

Fig. 2.2 p. 23 Cross from Lanherne, Cornwall (photo. Woolf/Greenham 

Fig. 2.3 p. 25 Pillar-stone from Welltown, Cornwall (photo. Woolf/ 
Greenham Collection). 

Fig. 2.4 p. 26 Pillar-stone from Lewannick, Cornwall (photo. Woolf/ 
Greenham Collection). 

Fig. 2.5 p. 30 Chape from St. Ninian's Isle (Royal Museums of Scotland). 

Fig. 3.1 p. 38 Location of Birsay and sites around Birsay Bay. 

Fig. 3.2a p. 43 Buckquoy: plan of phases I-II. 

Fig. 3.2b p. 44 Buckquoy: plan of phases H-VI. 

Fig. 3.3 p. 45 Red Craig (Area 3). 

Fig. 3.4a p. 45 Red Craig (Area 5). 

Fig. 3.4b p. 46 Red Craig (Area 5). 

Fig. 3.5 p. 48 Brough of Birsay (site IX) Phase 1 structures 13, 14 & 15. 

Fig. 3.6 p. 49 Brough of Birsay (site VII) Phase 1 structures 19 & 20. 

Fig. 3.7a p. 52 Brough Road south of Red Craig (Area 1) Cairn 2. 

Fig. 3.7b p. 52 Brough Road south of Red Craig: cist-grave exposed below 
Cairn 2. 

D <fkq e 

Fig. 3.8 p. 54 Brough of Birsay (Site IV South): stone with ogam inscription. 

Fig. 3.9 p. 56 Saevar Howe, Birsay: drawing of iron bell. 

Fig. 3.10 p. 62 Tintagel, Cornwall: aerial photograph. 

Fig. 3.11 p. 63 Gateholm, Dyfed: plan. 

Fig. 4.1 p. 82 Map of distribution of Phocaean Red Slipware in western 

Fig. 4.2 p. 83 Relative frequency of imported E ware on Insular sites. 

Fig. 4.3 p. 86 Reconstructions of glass vessels with opaque white trailed 
decoration from Dinas Powys and Whithorn Priory. 

Fig. 5.1 p. 94 Map of place-names containing cill. 

Fig. 5.2 p. 94 Amended map of place-names containing cill. 

Fig. 5.3/?. 96 Map of place-names containing cill and both (with religious 


The second Day Conference on Scotland in the Dark Ages was held on 18 
February 1995, two years on from the first, which looked at Scotland in a 
European context. This time the focus was more narrowly on the British Isles 
in an attempt to see Scotland as part of Celtic Britain-although there was a 
regrettable absence of input from Ireland. As previously it was intended to 
make the tone of the Conference inter-disciplinary, and the mix of papers once 
more provided a stimulating array of different and diverse approaches. Such 
a recipe certainly seems to provide an acceptable and attractive programme for 
students and teachers from varying departments, as well as for members of the 
general public: all of whom are well able to listen to and learn from the 
presentation of very different research topics by historians, archaeologists, 
linguists and onomasticians. Once more the numbers prepared to travel to 
St. Andrews for such an occasion in the middle of winter surprised and 
delighted the organisers. 

Some of our speakers also travelled a fair distance and we were exceedingly 
grateful to them for being prepared to down tools in the middle of a teaching 
term and devote time and energy to presenting papers and participating in the 
event. They are the reason for the success of these occasions which will be clear 
from the publication of their papers in this volume. All of them produced their 
disks promptly and helped to make my j ob as editor that much easier. The value 
of the day' s proceedings is too precious for the spoken word not to reach a wider 
audience. However unique an experience it is to hear the masters and 
mistresses of their subject expound in person there is a duty to catch the gems 
and fix them in precious metal (the printed word!) for their worth to be fully 

The papers are printed in the order in which they were delivered. Wendy 
Davies started the proceedings off with a very wide-ranging look at the idea of 
'protected space' and the concomitant practice of ecclesiastical sanctuary in 
early medieval societies. This opened up the various concepts which are 
evident in Scotland from the Germanic 'girths' to the 'comraich' of Gaelic 
society (as well as other terms). These can only be understood in the context 
of the practices of the early medieval church elsewhere. From concepts we 
moved to visual reality with Elisabeth Okasha's study of the carved and 
inscribed stones of south-west Britain. She usefully drew comparisons 
between these and the inscriptions of Pictland, andpointedoutthatthe evidence 
might suggest a rather higher level of literacy in Pictland than in the south-west. 

The next two speakers made the archaeological contribution to the day's 
events; Christopher Morris brought the situation of 'Dark Age' power centres 
into focus and presented his case for seeing Birsay in Orkney and Tintagel in 
the west country as secular residences rather than monastic retreats. Despite his 
opening comments on the unsuitability of the term 'Dark Age ' for the centuries 
of the post-Roman era it serves a function for me in encapsulating the problems 
of the period with which this volume and its predecessor are concerned. This 
is aperiod in which historical sources are fragmentary, terse and difficult to link 
into any coherent political narrative; for which reason those of us who are 
historians need to look to the discipline of archaeology, to place-name studies 
and to sculptural analysis for illumination. 

Ewan Campbell amplified the role of such power centres in Dark Age society 
by presenting the material evidence for wider contacts and by pointing out the 
different explanations which can be postulated about trade in such societies. 
The next two papers presented evidence about the ecclesiastical situation; 
Simon Taylor shows how place-names can be used to attempt to fill out the thin 
record of the church' s development in eastern Scotland. Thomas Clancy helps 
us to understand the important place which the Celi De reformers had in 
Scotland in the ninth and tenth centuries by demonstrating the close link 
between Iona and the Celi De reformers in Ireland, and the role of one of the 
abbots in transmitting the new reforming ideas eastwards to the heartland of the 
Scotto-Pictish kingdom. 

Finally Patrick Wormaldrangedmagisteriallyoverthe whole world ofEuropean 
'Dark Age' hegemonies while exploring the remarkable change of political 
power structures in eastern Scotland which is one of the central mysteries of this 
period. He sees this as a possible example of the violent take-over of power 
which is manifested elsewhere in Celtic Britain and Germanic Europe, and he 
puts forward another angle on the Viking role in these turbulent events of the 
ninth century. 

I believe the world of scholarship is enlightened by this collection of diverse 
studies which I am proud to present to the reading public, and hope that it will 
help to foster a better understanding of the culture and identity of this country. 
My thanks are due to the many supporters who have helped me to realise these 
gatherings; the Committee for Dark-Age Studies in the University of St. 
Andrews and the Early Medieval Research Group in Edinburgh (EMERGE) in 

2 Bfi&n 

particular. Professor John Guy, Chairman of the School of History and 
International Relations in the University provided a warm welcome and 
alcoholic cheer at lunchtime (which was much appreciated) as well as financial 
contributions towards Speakers' expenses, as also did the Russell Trust. 
Graeme Whittington provided the entree to the Dept. of Geography again as 
well as much practical help on the day. Dr. Ronald Cant as always was generous 
with moral and financial support in the realization of my Dark Age Studies 
initiative and the second of the ensuing Conference publications. 

Dept. of Medieval History 
University of St. Andrews 

S Britain. 

'Protected Space' in Britain and Ireland in tke Middle Ages 

Wendy Dalies 

This is a discursive and speculative paper, intentionally so, since I aim to ring 
bells in your heads and encourage you to think about possibilities. It does, 
however, take its starting point from a detailed analysis that I have discussed 
elsewhere (Davies 1995). 1 

The notion that some clearly defined spaces are places of protection, places 
where people can be safe, is a familiar one. It is probably best exemplified by 
the Christian idea of sanctuary (still invoked, for example, by illegal immigrants 
to Britain) but it is in no way confined to Christian contexts : there are the ancient 
Greek sanctuaries associated with Diana at Ephesus or Minerva at Sparta, for 
example, and there are the Levitical cities of refuge of the Old Testament 
(Timbal Duclaux de Martin 1939: 18-19), while the idea remains common in 
children's play; even the despised Violet Elizabeth had to be protected by 
William Brown and his friends when she 'took sanctuary' with them. 

In this paper I want to draw your attention to the act of protecting as well as 
to the places where people could feel protected, although - as we shall see - the 
two came together in the later middle ages. My comments will relate to Britain 
and Ireland, and largely to the central middle ages, but I will make some 
reference to later periods. 


The idea that an individual can legitimately afford protection to another was a 
norm in early medieval European societies, continental as well as insular. By 
'legitimately' I mean that it was a power, and a right, acknowledged by the 
society surrounding the individual and in several cases articulated in legal 
collections; Anglo-Saxon law codes offer a good insular example. It was usual 
for a man' s power to protect to be related to his status : the higher his status, the 
greater that power. This finds its most concrete expression in the graded 
penalties for breach or violation of protection that feature in many early 
medieval legal texts: if you killed a man in a bishop's house, the penalty was 
greater than if you killed him in a priest' s house; the offence to the bishop was 

greater, because his status was higher; and you owed compensation to the 
bishop for violating his protection in addition to the compensation you owed 
for the deed itself to the dead man's kin or lord. 

So much is standard; it is familiar and well-known. Before going further, I 
will take a few minutes to consider words for 'protection' and their semantic 
ranges, since they are important for our understanding of early medieval 
approaches. In Germanic languages the word for protection is mund and in 
Celtic languages there are words related to the root snad-, words like modern 
Welsh nawdd, Old Irish snddud, Old Breton nodet. The meaning at the core of 
mund is 'hand', and hence its secondary sense of 'protection', taking into the 
hands of someone. At the core of snad- seems to be the meaning 'to bind', and 
hence its sense of being attached to someone. Both groups of words were used 
to express the reciprocities that underlie the power and responsibility of the 
head of a household in the early middle ages: he protects and guards the- 
members of the household, and so they are in his hands and are thereby bound 
to him; they are in his power; as protector, he receives compensation for 
offences committed against them. The seventh- and eighth-century Lombard 
law on mundium (the sphere of someone's protection) is prominent: for 
example, a widow may return to the mundium of her near relatives, from that 
of her husband' s relatives, in certain circumstances, and a girl can return to the 
mundium of her relatives if he who holds her mundium attacks her (Fischer 
Drew 1973: 79-80, 85-92, for example). Even in England in the early seventh 
century, the consequences of infringing protection were explicitly set out in the 
Laws of ^Ethelberht, 75 and 76 (Attenborough 1922: 14). 

By extension, therefore, the household was a 'safe' place; it was safe 
because it was protected. It was (or should have been) inviolable. 2 Attacks on 
or in a household struck at the householder' s power of protection and therefore 
struck at the heart of his own honour and identity. 'Household' here was often 
explicitly both house and the land around it: to early Icelanders safety extended 
to the land 'within the fence [round the house]' (Gourevitch 1987: 529); and in 
Irish secular law, the precinct round the dwelling was explicitly included in the 
inviolable area - maigen digona (Binchy 1941: 83). Valuing the inviolability 
of the household was not peculiar to the early middle ages: we still think of an 
Englishman's home as his castle; the 'homestead' is not just the place where 
you live but it is the place where you are safe (owning, as I do, a garden across 
which my neighbours have the only available access to their property, I am 


often conscious of a deep-seated sense of outrage when strange vehicles 
encroach; this is not merely proprietary: I feel threatened and fearful, I start to 
feel unsafe); and in the genre of the Western, the 'bad' men are always the ones 
who attack the 'good' homesteaders. In the classic film of the early '50s, Shane, 
whenever the bad men cross the stream and enter the protected area around the 
hero's house, the tension mounts -the music changes - we can read the visual 
signals - we know that the crossing is a threat to the safety of the hero's family. 

In the early middle ages, by a different extension,.a man might stretch his 
power of protection to people beyond the homestead. In England, as late as the 
early tenth century, a secular man could offer protection to a thief for a limited 
period, for three days if the protector was a thegn, for nine if he was a king (IV 
jEthelstan 6 (Attenborough 1922: 148, 150)); if violated, the value of his 
mundbyrd was due to the protector for breach of protection. In Ireland a 
protector could give legal freedom from distraint {snddud), a power for which 
there were elaborate, status-related rules: protection from distraint could last 
for fifteen days from one sort of noble {aire ard) but only for ten days from a 
lesser noble {aire tuiseo), just as the extent of the maigen digona also varied 
with rank. So intrinsic was the notion of protection to the fabric of Irish society, 
that these powers still existed in Ireland, and were exercised, in the late 
sixteenth century (Kelly 1988: 140-1). 3 

On the continent, by yet another early medieval extension, the word 
mundbyrd in Carolingian Francia came to refer to a special royal protection, 
given by the ruler to some selected individuals and to monasteries, such as 
Lorsch: the beneficiary was protected from all harm by the Icing; if he was 
harmed, then the guilty one was subject to pay a fine. This case involves an 
extension of the king's personal power of protection, an extension that 
reinforced his distinctively royal status (Ganshof 1965: 46). 

In the early middle ages the power to protect was an aspect of the status of 
free men. Its application says much more about the power, privilege and 
independence of the protector than it does about the status or condition of the 
individual who came within the sphere of his protection. Hence, in our texts 
breach of protection attracts far more attention than respect for it. It did so 
because being able to sustain protective power was essential for the public 
recognition of status and therefore for the maintenance of social and political 
order in that world. 


Tke terrttorialteation, of protection, 

There are, of course, other types of 'safe place' than the household and the 
proximity of a protector; such are the meeting places protected by early Kentish 
law in England, or the safety of the Icelandic thing, or the quiet and order of an 
Irish court (airecht) meeting for judicial business - no anger or incitement was 
allowed. The reasons for these kinds of safety are different from the safety that 
derives from a protector and they do not primarily arise from a person' s status; 4 
rather, they are to do with a community ' s need to meet and conduct its business 
without fear of attack. Hence, characteristically, weapons had to be put down 
by those arriving at the meeting, and drawing a weapon when the meeting was 
in progress was a particularly serious offence. In early English law these 
concepts are more frequently expressed by the word fridr than by mund, where 
fridr is 'peace', 'security', 'freedom from fighting', and secondarily, 'truce', 
'agreement'; fridstow is therefore the 'peace-place', the refuge or place of 
safety. The ideas are distinct from the act of protection inherent in the notion 
of mund. 

The power to protect was not without limits, and - at least until the tenth 
century - there was a tendency to express its extent in temporal terms when it 
stretched beyond the household; hence, we find protection for thirty days, or 
whatever period was appropriate.^ What is so interesting in insular areas is the 
fact that in certain contexts the power to protect became territorialized beyond 
the homestead: power to protect could be expressed in relation to a defined 
space. Hence, in Welsh, nawdd, 'protection', gave the word noddfa, 'place 
(ma) of protection' . And accordingly offences committed in the noddfa brought 
compensation to the protector as well as to the damaged party. This is most 
explicitly set out in the southern Welsh, late twelfth-century, CyfnerthRedaction 
of the laws: the payments due for offences in the noddfa (outside the cemetery) 
were to be split 50:50 between the abbot and 'learned youths' of the church 
(Pryce 1 993 : 1 80). There are tenth- and eleventh-century cases of compensatory 
payments of this type being made to clerics - in material from the church of 
Llandaff, from the South East (Davies 1995: 138-40); and there are many 
Iwelfth-century references to the use and violation of specific noddfeydd (Pryce 
1993: 170-3). There are also examples of apparently comparable compensation 
being paid to churches in Ireland, as to Armagh for 'outrage' in the late tenth 
century (6 Corrain 1 978: 22). It is usually assumed that the termonn lands that 
surrounded Irish churches and monasteries marked the physical extent of 

protected space (Hughes 1966: 148-9; Lucas 1967: 203-4), although the 
termonn is more explicitly associated with refuge and with the limits of the 
'holy' in the seventh and eighth centuries (Doherty 1985: 56-9). Armagh, 
Clonmacnoise, Kildare and Scattery Island certainly had beatable termonn 
lands by the ninth century. Hogan (1910) listed 32 termonn names for early 
Ireland - often associated with a saint's name, like Termonn Brigte or Termonn 
Ciardin. Where these names refer to ecclesiastical space, the space may, as Ann 
Hamlin has suggested, have been marked out by cross-inscribed stones 
(Hughes and Hamlin 1 977: 80-1). 

Now, the territorial expression of protection has a very strong ecclesiastical 
flavour once it gets beyond the limits of the freeman's homestead. The idea of 
the protective power of the saints is particularly strong in Old Irish material - 
witness Colman's Hymn invoking John the Baptist as protection (snadud), the 
early poems from Iona invoking Colum Cille as protection {snadud) and the 
developed analogies of the saint as lorica, breastplate (Stokes and Strachan 
1903: ii.301; cf. Clancy and Markus 1995: 153-4, 170). Many of the examples 
of protected space for which we have good recorded evidence, like Llanbadarn 
Fawr and Llanddewibrefi in 1 109, relate to monasteries and churches. 6 

It could be that we should add to these Welsh and Irish cases the further 
ecclesiastical examples of the so-called 'chartered sanctuaries' of northern and 
westernEngland-Hexham,Beverley,Durham,Ripon, StBuryan, andPadstow. 
Beverley, like several others, had graded penalties for the violators of its Yk 
miles of protected space in the late middle ages, increasing as they approached 
the church (Cox 1911: 126-7); Durham and Hexham had similarly graded 
penalties at least by the late eleventh century - marked out by crosses (Hall 
1989: 426-7). St Buryan and Padstow were both noted for their 'privileged' 
sanctuaries (Olson 1989: 72, 79). These chartered or privileged sanctuaries 
look like Welsh protected space because they are significantly different from 
'ordinary' sanctuary as we find it in English churches. Their special status was 
known, and remarked upon by late medieval and early modern travellers like 
Leland, and some northern cases had special powers in relation to unemendable 
offences (Hall 1989: 433). The extent oftheir special space tended to be marked 
outphysically, often by crosses (four atBeverley andHexham, six at Wetherhal 
(Cox 1911: 128, 155, 175)); the areas were large- 1-2-3 miles in diameter; and 
penalties for offences within tended to be graded (as they were in Irish canon 
and Welsh secular law). Given the English legislation which territorialized 

£ Brvt&tn, 

protection in the tenth and eleventh centuries (see further below), we should 
perhaps expect them to be comparable. It does not follow that they had all the 
characteristics of the Welsh noddfa (in particular we do not know about the 
range of punishable offences in the crucial period of the tenth and eleventh 
centuries) but they are sufficiently similar to be considered in the same light; 
and sufficiently unlike sanctuary as it is commonly found. We should note that 
the areas around the salthouses in eleventh-century Nantwich and Middlewich 
were also especially protected: specified offences committed within a league 
of the two places attracted fines, although there were no penalties beyond the 
league {Cheshire Domesday, S2 (i.268r)). The latter material, which happens 
not to be ecclesiastical, provides us with clear evidence that protection was 
territorialized in practice, and not just in theory, in eleventh-century England. 

We should also think of the girths of Scotland, like that mentioned at Luss 
in 13 15 {RRS 5 : no. 55); Lesmahagow, Innerleithen, Tynninghame and Wedale 
church (Stow) all have much earlier references, although these earlier, twelfth- 
century, references use Latin words for the girth (Lawrie 1905:no. 172,^5 
1 : no.219, RRS2: no.68). 7 'Girth' is ametathesizedform of the late OldEnghsh 
word grid; where grictis a borrowing from Scandinavian grid; whose early 
semantic range (though not root) was very close to that of mund: it had apnmary 
meaning of 'home', 'abode', extended to 'peace', 'truce', 'pardon', and so on; 
it often meant 'safe conduct' in Old Icelandic (S0rensen 1993: 159). By the 
eleventh century it could mean in English both 'protection' and also a specific 
'peace' or 'truce', limited in space or time. The Oxford English Dictionary (s.v. 
grith) cites these meanings of peace and protection but also cites the use of girth 
for 'place of protection' (both general and specific) from 1300 until the 
nineteenth century. A high proportion of OED citations of this latter usage are 
in Scottish contexts. Not only do we find 'girth' as the word for ecclesiastically 
protected areas in Scotland; we also find grithcrosses at Tynemouth; and 
grithmen attached to northern English church areas, such as Ripon, m the later 
middle ages (the grithmen may or may not have sought protection - m 
Scandinavia grithmen were members of the household, whether permanent 
family members or servants on short-term contracts (Sorensen 1993: 159)). 
Just as Beverley and Tynemouth had crosses to mark the limits of the protected 
area, so Lesmahagow had four such crosses by 1 144 (Lawrie 1905: 136) and 
the monastery of Applecross in western Scotland had stone markers to lay out 
its protected territory - a place to this day known as a' Chomraich in Gaelic, the 
'protected place' (MacDonald 1985: 179). 8 

£ Britain 

As Professor Barrow pointed out at the conference, it may also be the case that 
Scottish placenames in tearmann - as in Tillytarmont and Drammietermont - 
point to Gaelic versions of the same phenomenon, given this use of the Irish 
word termonn; hence Clach an Tearmainn, the termonn stone, marking the 
limits of Oronsay Priory in the strand separating Colonsay and Oronsay 
(Watson 1926: 259). I am not confident, however, that all termonn names 
denoted 'protected' space in the sense used here, particularly in the earliest 
Christian centuries.^ 


Whatever its origin, the spatial dimension of extended protection looks largely, 
though not entirely, ecclesiastical; it is both a reflection of, and a contributor to, 
the status of specific churches. 

By the eleventh century space protected by ecclesiastical bodies in Britain 
and Ireland also looked like 'sanctuary'; and in part it was. But, it involved a 
lot more than sanctuary and the reasons for its existence were quite different. 
Firstly, the physical scale of insular protected space was altogether different 
from the 'ordinary' sanctuaries we find in Britain and from sanctuary on the 
continent. In the case of protected space, we are dealing with zones that could 
be as big as one, two or three miles in radius in Wales, Scotland, and England, 
as also in some Breton cases: St Asaph reputedly had a mile square (Pryce 1993: 
1 7 In); Applecross was twelve miles across (W atson 1 926: 1 25) and Luss three 
miles 'around' (RRS 5: no. 55); Hexham and Ripon were two miles across, 
Beverley three miles, and St Buryan and Padstow perhaps the same (Cox 1911: 
215, 223); and the original minihi of Gouesnou in western Brittany was about 
1.3 square miles, though later doubled (Tanguy 1984: 15). This contrasts with 
the classic 30-35-40 metres (passus) of sanctuary land around churches on the 
continent and with the tiny 'sauvetes' of southern France. 10 Secondly, the 
nature and scale of the consequences of violation of protected space were 
altogether different: if an offence was committed in a protected space (theft, 
abduction, arson, assault, homicide), then it occasioned significant financial 
compensation to the protector (as well as to the damaged), a compensation 
supported by secular law, as befitted an issue of personal status. Violation of 
sanctuary might well require penance, in recognition of the affront to the 
church, but that was essentially a spiritual matter and did not have to have 


material consequences; hence, excommunication was a common penalty for 
breach of sanctuary. Although the canonists began to discuss the application of 
secular penalties for breach of sanctuary in the eleventh century, Timbal 
Duclaux de Martin knew of no actual cases of secular penalties being applied 
at that time (1939: 207-8; 185, 237). 11 

This difference is hardly surprising given that sanctuary is in essence to do 
with asylum - refuge. It is a mechanism for ensuring the safety of life and limb 
for fugitives: at its heart is the legal and social position of the person seeking 
protection, not the status of the protector. From at least the time of the 
Theodosian Code, churches had provided a place of asylum for fugitive slaves 
and others (as the Levitical cities of refuge had provided asylum for homicides 
(Numbers 3 5)) (Timbal Duclaux de Martin 193 9: 83 -4). These ideas were quite 
clearly taken into Irish ecclesiastical thinking by the eighth century, for canons 
explicitly refer to Old Testament cities of refuge {Hibernensis XXVTII; cf. 
Doherty 1985: 57); as they also reached England by the late seventh century 
(Laws of Ine 5, Attenborough 1922: 38). Refuge for criminals could certainly 
be subject to restrictions: there were early medieval distinctions between the 
refuge rights of accused and condemned criminals. Nevertheless, sanctuary 
was essentially for fugitives, although ultimately the type of fugitive who could 
expect to be protected came to be severely limited (thieves and brigands were 
excluded in the thirteenth century and fifteenth-century papal bulls exempted 
further categories of offender). 

The restrictions of the later middle ages followed an extension of the scope 
of local sanctuaries on the continent in the years round about AD 1000, in the 
context of the Peace of God movement: there should be a special peace for a 
church and the houses around it, like the Catalan precincts known as sagreres, 
sacredplaces(HeadandLandes 1992; Marti 1988). What this meant in practice 
was a right to freedom from molestation for church property and often, given 
the prevailing political context, for the poor and defenceless or unarmed. In 
other words, it meant extending the right of asylum from fugitive criminals to 
the poor. The movement was taken to extreme lengths in some parts: in north- 
west Herefordshire (Leominster and neighbourhood) 'refuge cemeteries' were 
consecrated in the mid twelfth century, which had no associated rights of burial 
at all; these were places 'for the refuge of the poor in time of hostility ' - a sacred 
refuge without an associated holy focus (Kemp 1988: 86, 89); at the same 
period Bishop Stephen of Rennes blessed a cemetery 'for the refuge of the 

2. Drvtam 

living, not the burial of the dead' in the parish of Marmoutiers (Timbal Duclaux 
de Martin 1939: 230). Although, therefore, sanctuary was often expanded in 
western Europe in the eleventh and twelfth centuries, it was not expanded to 
become protected space after the sense of nawdd or mund. The idea of asylum 
remained central. 

Tke chronology of territorialization, 

This spatial expression of protection does not look especially ancient. I doubt 
that it startedmuchbefore the tenth century in Britain. Although the Hibernensis 
and other seventh-/eighth-century Irish texts go to some length to discuss 
marking out the bounds of holy places (XLIV), by the sign of the cross, they are 
more concerned with refuge (canonical sanctuary) than with protection (Lucas 
1967: 184; Doherty 1985: 56-7). 12 

My preference for the relatively late development of protected space arises 
for the following reasons. Firstly, it is extremely difficult to find strong and 
well-evidenced suggestions of the practice before about 900. Secondly, there 
is a coalescence of English cases of 'chartered sanctuary' attributed to the 
period of ^Ethelstan: rightly or wrongly people tended to believe that it was 
^Ethelstan who had confirmed or established the areas. Thirdly, the English 
legislation on mund became noticeably more territorial during the tenth 
century, and by the time of ^Ethelred and Cnut was strongly so: Alfred' s stress 
on periods of protection gave way to ^Ethelred's stress on the violation of 
protected space (Vffl ^Eth. 1 , 4) (cf. Hall 1989: 43 1); all churches were in the 
protection (grztfJofGod and Christ (VEIiEth. l,ICnut2(Robertsonl925: 116, 
154); while the penalties due to churches for violation of their protection varied 
since, though all had the same sanctity, all did not have the same status (VIII 
iEth. 5) - an important clause, emphasizing status and differentiating the 
authority that derives from status from the authority that comes with holiness. 
Fourthly, none of the Irish cases of compensation for infringement comes till 
the late tenth century. Fifthly, as I have argued elsewhere, there are special 
reasons for believing it to be a tenth-century development in Wales (Davies 
1995: 163-4). It is also likely that the change to a territorial approach was 
influenced by the - intrinsically territorial - canonical law of sanctuary. 
However, it may be as important that the change was a part of a general shift 
in attitudes to physical space: land became something to be delimited, ridden 

£ Britain. 

around, andphysically dominated, ramermanageneralizedsourceofsustenance 
or a distant source of goods to be plundered and vacated; property literally had 
to be encompassable. 13 

of <7ri<f into English. This is a tenth-century borrowing. In tenth- and eleventh- 
century English royal legislation gW<Twas used in parallel to mund: this is quite 
explicit in the Laws of Cnut, where 'rnundbreach' in Wessex is 'grithbreach' 
in the Danelaw (although the usages are not always so regionally distinctive) 
(II Cnut 12, 15; Robertson 1925: 180). In fact, by the year 1000 ciricgrith was 
a more common term than mund for the protection afforded by the church, 
whatever the region - and a more common term than the frid; peace, of the 
church though the terminology is not stable (by the eleventh and twelfth 
centuries Latin pax could refer both to a specifically given protection and to a 
general peace). It looks as if English ideas began to change round about 900: 
it has been argued that the Laws of Alfred started to equate church peace with 
house protection (Riggs 1963: 34); andin Edward and Guthrum's Peace (2.1) 
ciricgridvras to be as inviolate as the specific protection given by the king (the 
king's handgrid). The earlier 'peace' (frid) was becoming the 'protected area'; 
the protected area was more than a refuge for criminals (xe more than 
sanctuary)" penalties for violation of the ecclesiastical protected area m the 
tenth and eleventh centuries could be heavy; and perhaps some special 
protected areas, of considerable size, were marked out on the ground. 


We have come a long way from the simple sphere of a freeman' s protection, 
his 'home'. Quite different ideas like canonical sanctuary (the consecrated 
refuge) came to Britain and Ireland and influenced approaches to church 
territory- moreover, approaches to protected space continued to change and 
develop.' In England, Scotland and Wales, overtime, the protected space could 
become an immunity (an area exempt from the demands of others, such as 
demands for taxes or to inspect) or could even become a seigneune (an area 
which a lord set out to dominate, over which he pro-actively established 
control by exercising judicial powers and setting up monopolies). This could 
happen to the noddfa, though it did not necessarily do so. It is true of Llandaff 
by the early twelfth century, with its royal exemptions, and its rights to hold 

IS BrltiW 

courts, market and mint (Davies 1995: 150-2). It is true of Beverley and Ripon 
by the thirteenth century (and possibly even by the late eleventh century), with 
their extensive jurisdictional rights and powers (Lobel 1934: 126); it is true of 
the late medieval minihi in Brittany (Chedeville and Tonnerre 1987: 354, 358) 
and it is true of Luss by the early fourteenth century, with its rights of criminal 
jurisdiction (RRS5: no. 55). In other words, the protected space so contributed 
to the status of some particular institutions that it became the core of wider 
powers; and of the successsful attempts of some lords to turn power over land 
into power over people. 

Protecting space gave considerable financial and practical powers to some 
major religious bodies in the tenth and eleventh centuries - the period at the 
heart of the development. We do not need to explain these newly acquired 
powers in terms of the devolution and fragmentation of royal or imperial power 
(as historians are prone to do for the continent). Indeed, in some insular cases 
new ecclesiastical powers seemed to develop as a response to increasing rather 
than decreasing ruler power: in Wales it was a defensive reaction against ruler 
aggression; in England the church's power to protect became in part an 
expression of ruler power, for each reinforced the other. 

z[' -}c -Jc ^ 

It was my purpose in giving this talk to draw attention to protecting space 
in Britain and Ireland and stimulate some further thought. The phenomenon is 
a commonplace for historians of early medieval Ireland and Wales. It looks as 
if it may be as characteristic of England in the tenth and eleventh centuries, 
particularly in view of the framing of English legislation and the evidence of 
Domesday Book; and perhaps it may not be quite so characteristic of Ireland as 
is commonly assumed - at least at the early, seventh- and eighth-century, 
period. The same does not appear to be true of the continent, with its much 
greater emphases on ecclesiastical immunities. The use that insular ecclesiastical 
bodies initially made of protected space had some of the same consequences for 
them as the early immunity had for Frankish churches - additional income 
(Fouracre 1995); in political terms it was the functional equivalent of the 
Frankish immunity. 

What is significant about this institution is not the mere delimitation of a 
special space but the distinctive nature of the powers exercised within the 


u Br 


space It is these that differentiate protection from sanctuary. The subject could 
tolerate a much lengthier investigation, particularly with reference to what 
happened inside the English chartered sanctuaries, but also with reference to 

to the Scottish and Irish termonn - the word itself has a quite different range of 
reference from Welsh noddfa (borrowed from Latin terminus, boundary) and 
we should perhaps be looking at comrach rather than termonn for close 
parallels. I hope others will pursue the trail: not only does it look rich; but it lies 
at the heart of strategies for establishing and sustaining power, both personal 
and corporate. And that is central to our understanding of social and political 
development in the early middle ages. 14 

Appendix ■ — ■ 

A note on ecclesiastical immunities 

By the tenth century the continental immunity (a privilege which freed the 
holder of immune lands from various kinds of public intervention in those 
lands especially with regard to taxation and public systems of justice) was 
largely an ecclesiastical phenomenon - one that had points of contact with 
'protected space' . If the immunist took fines for offences committed within the 
immunity, this must have looked rather like the consequences of violation of 
'protection' However, although some control of judicial process might lie in 
thehandsofan ecclesiastical immunist lord, the process still in theory remained 

but outside the sanctuary (eg theft between lay persons) carried additional 
compensations to the ecclesiastical lord over and above the 'public' fine. The 
pre-twelfth-century immunity is therefore very different from 'protected 
space' In any case, at least in the Carolingian period, those guilty of major 
crimes had to be handed over to public authority by the immunist. And, further, 
we have reservations nowadays about the power of the immunist even in 
relation to minor offences (Fouracre 1995: 58-68). By the twelfth century 
canonists began to apply penalties for violation of immunity to violation of 

immunity less clear than it had been before (Timbal Duclaux de Martin 1939: 
147-58, 185-96); these are late developments, however, and do not affect the 
clear distinctions of the seventh to eleventh centuries. 


There is another type of immunity, sometimes called the 'narrow immunity' 
by continental scholars, which also has features which resemble insular 
protected space. The twelfth-century canonists argued that every church had a 
special immunity stretching for 30 (and in some cases 40) paces (passus); if 
anyone committed sacrilege by injuring or stealing from clerics within this 
area, then they were liable to pay a money fine (Gratian Decretum, C.17, q.4, 
c.21). Although the size of this zone was similar to the 30-40 paces of the 
sanctuary zone, the offence committed was quite distinct from violation of 
sanctuary; both were sacrilege, but damaging clerics near the church was a 
different kind of sacrilege. The special unmunity was therefore different from 
sanctuary right in canonist thinking. This special immunity was also 
differentiated from 'protected space' by its small size and by the absence of 
secular enforcement of its provisions; and, in any case, the extent to which the 
canonists ' theory was put into practice is very uncertain (and would repay some 

There were, then, two different kinds of immunity in canonist thinking, and 
each was different from sanctuary. I would argue that all three were different 
from pre-twelfth-century 'protected space', although the 'ordinary' immunity 
couldbe comparable in size and the 'special' immunity could in theory involve 
payment of fines to a church. The similarities are such that, despite the 
differences, it seems reasonable to propose that the Frankish immunity performed 
much the same function for continental churches as protected space did for 
insular churches; and that by the twelfth and thirteenth centuries protected 
space could easily become the core of an insular immunity. 15 

University College London 
2 July 1995 

Abbreviations ~~~~~ ~~~~ ~ , — ~~ — ~ 

Cain Adamndin: ed. K. Meyer, Anecdota Oxoniensia, Medieval and Modern 
Series 12, Oxford 1905. 

Le Cartulaire deRedon: ed. A. de Courson, Paris 1863. 

Cheshire Domesday: Domesday Book, vol. 26, Cheshire, ed. P. Morgan 
(general editor J. Morris), Chichester 1978. 


Decretum Magistri Gratiani: in Corpus Iuris Canonici, ed. E. Friedberg, 2 
vols., Leipzig 1879-81. 

Hibernensis: Die irische Kanonensammlung, ed. H. Wasserschleben, 2nd edn 
Leipzig 1885. 

RRS 1 : Regesta Regum Scottorum, vol. 1 , The Acts of Malcolm IV King of Scots 
1153-65, ed. G. W. S. Barrow, Edinburgh 1960. 

RRS 2: Regesta Regum Scottorum, vol. 2, The Acts of William I King of Scots 
1165-1214, ed. G. W. S. Barrow, Edinburgh 1971. 

RRS 5: Regesta Regum Scottorum, vol. 5, The Acts of Robert I King of Scots 
1306-29, ed. A. A. M. Duncan, Edinburgh 1988. 

Attenborough, F. L. (ed. and trans.) 1922, The Laws of the Earliest English 
Kings, Cambridge. 

Binchy, D. A. (ed.) 1941, Crith Gablach, Dublin. 

Chedeville, A. and Tonnerre, N.-Y. 1987, La Bretagne feodale XI e -XIII e 
siecle, Rennes. 

Clancy, T. O. and Markus, G. 1995, Iona. The Earliest Poetry of a Celtic 
Monastery, Edinburgh. 

Cox, J. C. 1 91 1, The Sanctuaries and Sanctuary Seekers of Medieval England, 

Davies, W. 1995, 'Adding Insult to Injury: Power, Property and Immunities in 
Early Medieval Wales' in W. Davies and P. Fouracre (eds.), Property and 
Power in the Early Middle Ages, Cambridge, pp. 137-64. 

Davies, W. and Fouracre, P. (eds.) 1995, Property and Power in the Early 
Middle Ages, Cambridge. 

»i BtrttAin 

Doherty, C. 1985, 'The monastic town in early medieval Ireland' in H. B. 
Clarke and A. Simms (eds.), The Comparative History of Urban Origins in 
Non-Roman Europe, British Archaeological Reports International Series 255 (i), 
pp. 45-75. 

Edwards, N. 1990, The Archaeology of Early Medieval Ireland, London. 

Fischer Drew, K. (trans.) 1973, The Lombard Laws, Philadelphia. 

Fouracre. P. 1995, 'Eternal light and earthly needs: practical aspects of the 
development of Frankish immunities' in W. Davies and P. Fouracre (eds.), 
Property and Power in the Early Middle Ages, Cambridge, pp. 53-81. 

Ganshof, F. L. 1965, Frankish Institutions under Charlemagne, trans. B. and 
M. Lyon, New York 1968. 

Gourevitch, A. J. 1987, 'Semantics of the medieval community: "farmstead", 
"land", "world" (Scandinavian example)' in Las communautes rurales, pt 5, 
Recueils de la societe Jean Bodin pour I 'histoire comparative des institutions 
vol. 44, Paris, pp. 525-40. 

Hall, D. 1989, 'The Sanctuary of St Cuthbert' in G. Bonner, D. Rollason, C. 
Stancliffe (eds.), St Cuthbert, his Cult and his Community to AD 1200, 
Woodbridge, pp. 425-36. 

Head, T. and Landes, R. (eds.) 1992, The Peace of God: Social Violence and 
Religious Response in France around the Year 1000, Ithaca and London. 

Hogan, E. 1910, Onomasticon Goedelicum Locorum et Tribum Hiberniae et 
Scotiae, Dublin. 

Hughes, K. 1966, The Church in Early Irish Society, London. 

Hughes, K. and Hamlin, A. 1977, The Modern Traveller to the Early Irish 
Church, London. 

Kelly, F. 1988, A Guide to Early Irish Law, Dublin. 

15 Britain 

Kemp B 1988 'Some aspects of the parochia of Leominster in the twelfth 
century' in J. Blair (ed.), Minsters and Parish Churches. The Local Church in 
Transition, Oxford, pp. 83-95. 

Lawrie, A. C. 1905, Early Scottish Charters, Glasgow. 

Le Moing, J.-Y. 1988, 'Toponymie bretonne de Haute-Bretagne', These de 
Doctorat, Universite de Rennes 2, 2 vols. 

Le Moing, J.-Y. 1990, Les noms de lieux bretons de Haute-Bretagne, Spezed. 

Lobel M D. 1934, 'The ecclesiastical banleuca in England' inF. M. Powicke 
(ed.), Oxford Essays in Medieval History presented to Herbert Edward Salter, 
Oxford, pp. 122-40. 

Lucas, A. T. 1967, 'The plundering and burning of churches in Ireland, 7th to 

MacDonald, A. D. S. 1985, 'Iona's style of government among the Picts and 
Scots: the toponymie evidence of Adomnan's Life of Columba', Peritia 4, pp. 

Marti, R. 1988, 'L'Ensagrerament: l'adveniment de les sagreres feudals', 
Faventia 10, pp. 153-82. 

6 Corrain D. 1978, 'Nationality andkingship in pre-Norman Ireland' inT. W. 
Moody (ed.), Nationality and the Pursuit of National Independence, Belfast, 
pp. 1-35. 

Olson, L. 1989, Early Monasteries in Cornwall, Woodbndge. 

Pryce, H. 1993, Native Law and the Church in Medieval Wales, Oxford. 

Riggs,C.H. 1963, CriminalAsyluminAnglo-SaxonLaw,\Jmversity of Florida 
Monographs 18, Gainesville. 

Robertson, A. J. 1925, TheLaws oftheKings of England from Edmund to Henry 
I, Cambridge. 


S0rensen, P. Meulengracht 1993, Forta=lling og cere. Studier i 
islcendingesagaerne, Aarhus. 

Stokes, W. and Strachan, J. (eds.) 1901-3, Thesaurus Palaeohibernicus, 2 
vols., Cambridge. 

Tanguy, B. 1984, 'La tromenie de Gouesnou. Contribution a l'histoire des 
minims en Bretagne', Annales de Bretagne, 91, pp. 9-25. 

Timbal Duclaux de Martin, P. 1939, Le droit d'asile, Paris. 

Watson, W. J. 1926, The History of the Celtic Place-Names of Scotland, 

1 I owe thanks to Professor Richard Bailey, whose comments in a coach on a 
trip to Luss first stimulated me to think about these issues. 

2 Note that even in late eleventh-century Cheshire offences in the homestead 
- in domo - carried a special fine (Cheshire Domesday, C4 (i. 262v)). 

3 Violation of the legal protection of a freeman by killing or injuring someone 
under protection was the offence known as diguin in Irish law. Binchy (1 94 1 : 
83) points out that the meaning of diguin extended to include 'breach of house- 
peace', where the 'house-peace' encompassed the surrounding precinct as well 
as the house. 

4 Although, if a king presided at a meeting, it would of course reinforce the safe 
status of the occasion. 

5 In Ireland, at least, the limits also included certain types of person; some could 
never be protected - a runaway slave, a runaway wife, a killer, a son who failed 
to look after his father; this clearly distinguishes the rules about protection from 
those about sanctuary (Kelly 1988: 141). 

Dat?k<\ „„ 

6 Breton minihi (< Latin monachia, monastic land) may have been used in a 
fasMonsimilarton 0 ^/fl,almoughits usages were clearly various (CMdevdle 
andToraierre 1987' 354-8). The vernacular term was in use aheady mthemia- 
nintb. century (Cartulaire de Redon nos. 141, 142, 181, 193), although it is not 
clear that it had any meaning at that period beyond 'monastic land - some ot 
which was aheady in lay hands. From the later middle ages the term meant 
sanctuary, in the conventional sense (Tanguy 1984: 24). The place-name 

1988- ii 78- LeMoing 1990: 234; Tanguy 19 84). Most frequently of parish size 
(eg Gouesnou, Locronan), a few minihi were enormous (more than eleven 
parishes round Treguier) and some were very much smaller than the pansh (eg 

Gouezec); they were sometimes marked out by crosses. Given the parish size 

meaning of ecclesiastical protected space in the tenth and eleventh centuries; 
the Life of St Goulven emphasizes the inviolability of the saint s space at 
Goulven, Finistere (Tanguy 1984: 21-2). 

7 I am extremely grateful to Professor Geoffrey Barrow for supplying me with 
references to girths and tearmann names after the February conference. 

8 F 0 rtheanglicizedIrishwordc 0 mri C fc, 'legal protection', which was used in 
English official documents in Ireland in the late middle ages and early modern 
period see Kelly 1988: 141n; alsoBinchy 1941: 107, on different Irish words 
for 'protection', (Irish commerce, and other words for protection, seemtohave 
replaced snddud in the later middle ages.) 

9 It is also unlikely that monastic valla, earthworks delimiting an area around 
a monastery, as identified in Ireland, were to delimit 'protected space : .they 
define much smaller areas than the protected spaces we know about (Edwards 
1990: 106-12). 

10 However, unusually, St Denis claimed a large area (Timbal Duclaux de 
Martin 1939: 160). 

11 Of course, continental ecclesiastical immunities may have looked like 
'protected space' in the insular sense in the tenth and eleventh ce atones but 
they were different in their operation (Davies and Fouracre 1995: 12-16, 256- 
8); see further, Appendix. 



lz There is a germ of the protective notion in Hibernensis XLTV.8, with its 
warnings against the violation of holy places by homicide and theft, but the 
starting point is different - pollution of the holy -and at this stage the penalties 
appear to be purely ecclesiastical (ie penance). There is another germ of 
comparability in that penances were graded in accordance with the status of the 
sanctuary as well as the nature of the offence - a recognition of the importance 
of status (cf. Doherty 1985: 57-8). Within another couple of centuries some 
churches were claiming a right to secular penalties (dire) - not just penance - 
for offences committed by laymen within the termonn, enforced by secular 
rulers (Cain Adamndin, 3 6) . Though the conceptual approach is different here, 
the application of such claims must have had effects which were indistinguishable 
from the effects of violation of protected space; in other words, when such 
claims came to be acknowledged, the termonn became like the noddfa. 

13 There were strong feelings at the conference that the Irish and western 
Scottish development must have been earlier. The Irish development could of 
course have been precocious; but the signs are that (i) the Irish termonn in the 
seventh and eighth centuries essentially marked out a place of refuge, safety 
and safe-keeping, as it continued to do in the central and later middle ages 
(Lucas 1967); (ii) some of the termonn lands took on the attributes of a 
protected area in the ninth/tenth centuries (see also above, n. 12). Apart from 
the claims of seventh-century Armagh to an unbelievably large area within its 
terminus, we have little idea of the size of the termonn in the early middle ages. 
All this would benefit from a much closer and more systematic examination - 
and a closer look at what happened inside a termonn. 

14 I understand that Dr Brian Golding, of the University of Southampton, is 
currently working on a book on English sanctuary; this will be a very welcome 
addition to the existing literature. 

15 I am very grateful to my colleague David d'Avray for assistance with 
canonist texts. 

i§ Brltatn, 

Darl^ e 
19 Britain 


Tke Early Ckristian Carded and Inscribed Stones of soutk- 
west Britain 

Elisabetk Okaska 

In this paper the early Christian stones of south-west Britain are described and 
illustrated and some comparisons are then made with the Pictish inscriptions of 
Scotland. The texts ofthe inscriptions discussed and the system of transliteration 
employed are given in the Appendix. For a full discussion ofthe south-western 
stones see Okasha 1993 and ofthe Pictish (non-ogham) stones see Okasha 
1985. Five ofthe inscriptions discussed are illustrated in the accompanying 
figures and illustrations of all appear in one or more of Allen and Anderson 
1903, Okasha 1985 and Okasha 1993. 

The term ' south-west Britain' is used to refer to the south-west peninsula of 
England, that is, to the counties of Cornwall and Devon. This term, rather than 
for example 'south-west England', is used since the Cornish (though not those 
from Devon) regard themselves with some historical justification as Celtic not 
English. In part of the Romano-British and early Christian periods, in the 
fourth, fifth and sixth centuries AD, this area did indeed comprise the Celtic 
kingdom of Dumnonia. Exact details ofthe fate of Dumnonia are sketchy, but 
the general drift is clear: from the seventh century onwards the Anglo-Saxons 
fought their way into Dumnonia and pushed the Celts further south and west. 
By the late tenth century the whole area was under some sort of Anglo-Saxon 

In the context of carved and inscribed stones, the term 'early Christian' is 
used to refer to the entire period from the withdrawal of the Roman legions from 
Britain up to the Norman Conquest, in round figures from AD 400 to AD 1100. 
'Early Christian' is thus a term of chronology and does not imply that all the 
stones were erected by or for Christians. Some of them certainly were, those 
which consist of or contain a cross, for example, or those including a Christian 
symbol like the chi-rho. Some of the rest probably were, but we cannot be sure. 

There are 69 inscribed stones from this period still in existence and a further 
ten, though now lost, are known to have existed in modern times. This number 
may gradually rise. There is, for example, a newly found stone from Kenidjack 
near St Just in Penwith which contains a letter S followed by two letters, 

ii Britain 

nossibly an F and a C. Charles Thomas has recently suggested that this is a 

text may be better considered as later graffiti, perhaps someone s initials. 
However the general point remains valid that new inscriptions andnewpieces 
of sculpture are likely to turn up in the course of archaeological and other 

slabs orpillar-stones. The crosses are of course Christian monuments and there 
t ML of crosses in Cornwall dating from the ninth to the eleven h 
centuries, although most of them are not inscribed. Absolute figures are hard 
to give At present there is being undertaken a pilot survey of all the pre- 

are 25 sculptured stones which are reasonably certain to date from the pre- 
Conquest period. A cross from Sancreed, for example, (Fig. 21) — > a 

crucifixion figure with a halo, set 

1 ^ inside a beaded cross-head; beneath 

are panels of interlace, now rather 
worn, and an inscription. The right 
hand side contains a serpentine 
creature and the left hand side has 
diagonal key pattern. On the back 
there is interlace onboth the head and 
the shaft. The interlace carving 
suggests that the cross can be dated to 
the tenth or eleventh centuries. 

Much less easy to date are those 
crosses that contain only a crucifixion 
figure and no other ornament. There 
are 28 such crosses in Penwith. Some 
of these crosses may belong to the 
early Christianperiod, but it is hard, in 
the absence of ornament, to decide 
which ones are early and how long the 
tradition of erecting them continued. 

Fig. 2.1 Cross from Sancreed, Cornwall 
(photograph Royal Commission on Historical 
Monuments of England) 

2 i BrvtcW 

Then there are at least 1 00 uncarved, uninscribed crosses and cross-bases in the 
area. It is virtually impossible to put a date on crosses of this sort. Some may 
date from before the Norman Conquest, others may date from the early 
medieval period, some might even be from modern times. 

The preliminary figures for Penwith are: 25 stones with pre-Conquest 
ornament, 28 with crucifixion figures only, and around 100 with no carving. If 
Penwith is typical of south-west Britain, or even just of Cornwall, we are clearly 
talking of hundreds of pre-Conquest monuments, but of how many hundreds 
is not altogether clear. 

Only eleven of the actual crosses are inscribed although there are also three 
cross-bases that may have had inscriptions and there are three inscribed stones 
which were subsequently re-cut to form crosses. The inscribed crosses usually 
have the text integrated into the design of the carving, often in a panel of its own. 
The texts usually read horizontally and the lettering is most often in a 
predominantly insular, as opposed to a predominantly capital, script. The cross 
from Lanherne in Cornwall (Fig. 2.2) 

is one of the best-preserved of the 
inscribed crosses and it exhibits all 
these features. The text on one side 
reads RUHOL, presumably a personal 
name RUNHOL, a name that is likely 
to be Celtic. Runhol may have carved 
the cross or have commissioned it to be 
carved. The text on the other side is 
quite legible but is hard to interpret. It 
reads BREID [ET] [IJMAH and may 
consist of two personal names joined 
by ET, though this conclusion is based 
on little but the well-known epigraphic 
principle that if something is 
incomprehensible itmustbe apersonal 
name. If these are personal names their 
etymology is uncertain. 

As well as eleven inscribed crosses, Fig _ 2 .2 Cross from Lanherne, Cornwall 
there are two inscribed altar slabs. The (photograph Woolf/Greenham Collection) 

?? Britain 

one from Camborne in Cornwall is in use as a side ^^^T^C 
parish church of StMartin and StMeriadoc. Its text reads LEUTUT IUSIT HEC 
ALT ARE PRO ANTMA SUA, 'Leuiut ordered this altar for his own soul .The 
Latin is irregular in spelling, grammar and syntax: IUSIT for iussit is followed 
not by an infinitive but by a direct object HEC ALTARE. The ^classical altana 
(neuter plural) occurs, as is common in medieval Latin, as ALTARE and so 
HEC is presumably neuter and accusative, though whether it represents hoc 
(singular) or haec (plural) is unclear. The altar slab is dated to the tenth or 
eleventh centuries on the evidence of its T-fret carving. LEUIUT is probably 
a Celtic name, though it couldpossibly be English; there is for example a name 
leuiet in Domesday Book which has been explained as a spelling of the Old 
English name leofgeat. 

These crosses and altar slabs, containing interlace and other designs, may be 
crude by the standards of contemporary sculpture from these islands. They do, 

are aesthetically pleasing. They are also indubitably Christian and they can be 
dated within reasonable limits, probably from the ninth to the eleventh or 
twelfth centuries. When we turn to the third category of inscribed stones, the 
memorial stones known as pillar-stones, the position is rather different. Firstly, 
there are many more inscribed pillar-stones than there are inscribed crosses or 
altar slabs; there are 51, of which two are now lost. Secondly, they are typically 
pieces of uncarved granite whichoftenseemtohavebeenleftinthecrudeshape 
in which they were picked up from an outcrop of rock. Sometimes however, 
theyhavebeenrudely shaped into pillars and sometimes the inscribed surfaces 
have been roughly dressed. Thirdly, many of them lack any specific indication 
of Christianity. Fourthly, they are difficult to date with any precision and some 
can only be dated within the broad limits of the fifth or sixth to the eleventh 

The stone from Welltown near Cardinham in Cornwall is a typical piUar- 
stone (Fig 2 3) The text is set without any margins or panels and reads 
vertically downwards with the bottoms of the letters to the viewer' s left. The 
text reads [VA]ILAT[H]I [F]ILI VROCH[ANI], probably to be interpreted 
'(the stone) of [Va]ilat[h]us, son of Vroch[anus]\ This formula is frequently 
encountered. The names are Celtic, probably specifically Irish. The script used 
is predominantly capital with an occasional insular form. The only carving ; is 
the incised arc above the text and there is no explicit sign of Christianity. It is 




likely to date from the fifth or sixth 
to the eighth century. 

Several of these features can be 
observed again on the stone from 
Southill. Like the Welltown stone, 
this stone is fairly untypical in 
having a legible text, but it is quite 
typical in that the text is in a 
predominantly capital script, is set 
withoutpanels ormargins andreads 
vertically downwards facing left. 
The text reads CVMREGNI FILI 
MAVCI, '(the stone) of 
Cumregnus, son of Maucus ' where 
the name CVMREGNI is Latin 
and MAVCI is probably Celtic. 
Above the text is a double arc and 
above it a cross, or perhaps a 
monogram chi-rho, indicating that 
the stone was erected by or for a 
Christian. This stone exemplifies 
two common epigraphic features 
of pillar-stones : the use of the letter 
I set horizontally and the use of a particular form of the ligature F/I. These 
features are confined to inscribed stones from the south-west and from Wales, 
with the one exception of horizontal I used twice on a stone from Santon, Isle 
of Man (Nash- Williams 1 950, 11). The Southill stone probably dates from the 
sixth to the eighth century. 

One other sort of text occurring on the pillar-stones is the Christian formula 
hie iacet 'here lies', often spelt IC IACIT. The use of this formula implies of 
course that such stones were not just memorial stones but specifically grave- 
stones. An example occurs on a stone from St Just. The monogram chi-rho is 
incised on one face, neatly set within margins, while on one of the sides is the 
text SELVS IC IACIT, ' Selus lies here' , with the last I horizontal. There are two 
smaller letters, probably reading NI, above the name SELVS and these may 
indicate a spelling correction; that is, the name could be either SELVS or 

W Britain 

Fig. 2.3 Pillar-stone from Welltown, Cornwall 
(photograph Woolf/Greenham Collection) 

SENILVS, both Latin names, or 
SELNTVS, which would be more 
difficult to'explain. This stone, like 
the previous one, is likely to date 
from the sixth to the eighth century. 

Afew of thepillar-stones contain 
texts in both roman and ogham 
script. The use of ogham script is 
one of the features which suggests 
an early date for a stone, from the 
fifth or sixth century to the eighth 
century. An example of a stone 
with ogham is one of the ones from 

Lewannickinnorth Cornwall (Fig. 

2 4) It has a text in capitals which 

probably reads [HIC] IACIT 

VLC[A]GNI, ' [here] lies (thebody) 

of Ulc[a]gnus' . The personal name 

is Irish and occurs also on another 


Nanscow in Cornwall. On the 
Lewannick stone, the two ogham - - - 

texts are also both renderings of 

top down, one <**£?£™ ° was commem orated three times on this 
stone. It seems bkely that one person w memorial. 

The Lewannick stone therefore appears to be bi literal, wi v 
repeated in a different script. 

thatfromFardelmDevon,nowmmeBritishMuseum ^ 

because the language of its texts capitaltext 
Trish It contains four texts, two in capitals and two in ognam f 

a vMCim MAOVIFJNI and one ogham text reads MAQIQICI These 
reads FANON1 MA^ v ^ o , ■, t ^ specifically 

texts couldbeexplainedashavmgtlueeCelti« 

Irish; alternatively they could contain the early Irish word MA^ 

Fi g 2 4 Pillar-stone from Lewannick Cornwall 
(photograph Woolf/Greenham Collection) 

w Brltau 


meaning 'son', followed by personal names. The other ogham text reads 
SAFFAQQUCI, probably a Celtic name. The second capital text probably 
reads G[A]G[R]A[NV]I, presumably a personal name though of obscure 
etymology. The Fardel stone thus appears to commemorate several different 
people and its texts are not bi-literal in the sense that they do not repeat parts 
of each other in a different script. If some of these texts are in the Irish language 
then this stone is unique amongst south-western monuments in this respect. 

There is one other stone from south-west Britain which is unique on account 
of its language. This is the stone from Lanteglos in Cornwall which has a text 
in English. The text is now difficult to read and it has deteriorated since I first 
examined it in 1964. Using readings made in 1964 and 1985, a text can be 
reconstructed: AELSEL[D] 7 GENE [RED] [W]0[H]TE PYS[N]E 
be translated 'Aelsel[d] and Gene[red] made this ?farnily-place (or place of 
peace) for Aelwyn' s soul and for ?themselves (or Hey[sel]) ' . The text contains 
some difficulties but certain things are clear. The language is clearly English 
(not Latin or a Celtic language) and is late in date, probably eleventh-century 
at the earliest. Features typical of English of this date are the lack of inflection 
on S[0]UL, the loss of r in the word [W]0[H]TE 'made', and the spellings of 
the personal names, spellings which can be paralleled in Domesday Book. 

It has to be admitted that this stone is rather an embarrassment. Its language 
clearly proves that it cannot be dated to earlier than the eleventh century. The 
date-range of the pillar-stones has therefore to be extended to include the 
eleventh century. If the Lanteglos stone did not exist, one could terminate the 
series about two centuries earlier. Some people do so: Charles Thomas, for 
example, dismisses the Lanteglos stone as a 'fossil' (Thomas 1994, 327) and 
dates the series of pillar-stones to the fifth to the seventh centuries. To ignore 
the inconvenient is of course one way of dealing with the evidence; another way 
is to put forward a model that accommodates it. It seems to me that there is no 
alternative but to admit that pillar-stones were still being erected in the eleventh 
century. In this context we should note that in Wales there are two Class I stones 
dating from the eleventh or twelfth centuries, although the rest of this series of 
stones are dated no later than the ninth century. The two stones are those from 
Heneglwys and Llangors (Nash-Williams 1950, no. 5, p. 53 and no. 60, p. 76. 
See also Dark 1992, 60). 

The position regarding the early Christian inscribed stones of south-west 

Z J BirviSSn, 

Britain can be summarised thus: there are 79 inscribed stones, ^eluding ten 
nowlost, ranging in date from the fifth or sixth to the eleventh centunes These 
inscribed stones are mainly pillar-stones though some are crosses and other 
monuments. Roman script, either capital or insular, is used and there are six 
stones which have in addition an ogham text. The language used is generally 
Latin although there is one stone which has an English text and one which may 
havelrishones.Manytexts commemorate individuals and the majonty of these 
neople have Celtic names; sometimes we canbe more precise and say that the 
personal names are specifically either Irish or Welsh/Cornish, sometimes we 
cannot. The non-Celtic names are either English or Latin. 

Someusefulpoints of comparison can be made between these stones and the 
inscriptions fromPictland,thatis,matareaofScotlandwMchwasunderPictish 

control during the seventh, eighth and early ninth centuries. Included here are 
inscriptions from the area in the west which was intermittently under Pictish 
domination in this period. To facilitate comparison with the stones of south- 
west Britain, the position is first summarised and then discussed m more detail. 

There are some 45 Pictish inscriptions, ranging in date from the seventh to 
the ninth centuries. The inscriptions on stone occur on a range of monuments, 
pillar-stones, cross-slabs and, in particular, Class I and II symbol stones; mey 
also occur on objects other than stones, for example the St Niman s Isle chape 
and a knife handle from North Uist. Only one script, either roman or ogham, is 
used on any one stone, except for one rather doubtful stone from Newton 
Aberdeenshire, which might contain both. The languages used are Latin and 
Pictish and the personal names are Pictish, early Gaelic or Latin. 

The first point of comparison that can be drawn concerns the actual number 
of artifacts. In both areas, the number of inscribed monuments compared to 
those uninscribed is quite small. However, while the 79 inscribed stones of 
south-west Britain cover a period of six or seven centuries, the 45 or so Pictish 
inscriptions cover only two or three centuries. There are, that is, considerably 
more surviving inscriptions per century from Pictland than from south-wes 
Britain Of course the number of surviving monuments may reflect not 
numbers of original monuments so much as things like the subsequent use of 
the land, the density of population on it during the last millennium, and the 
amount of archaeological investigation that has taken place there in the last 
century However, these three factors, later land-use, population density and 

?J Britain, 

quantity of archaeological investigation, are probably broadly similar between 
thetwo areas under discussion. This would not be the case if, for example, either 
area were compared to Anglo-Saxon England. 

The numbers of surviving monuments suggest that literacy was more 
widespread in Pictland than in south-west Britain. Another piece of evidence 
suggesting the same thing is the wide range of Pictish monuments and artifacts 
containing inscriptions. In the south-west, inscriptions occur mainly on 
unsculptured pillar-stones with a few on other stone monuments like decorated 
crosses and altar slabs. In Pictland there are inscriptions on various sorts of 
stones and also on other objects. The ninth-century stone from St Vigeans, for 
example, is a cross-slab. It has interlace carving above the text and further 
carving on all sides, including Pictish symbols on the back. The text is a 
memorial one and probably reads DROSTEN IRE UORET [E]TT FORCUS. 
It contains three personal names, the first two being Pictish and the third Gaelic. 
The names are joined by [E]TT, presumably Latin et, and by IRE (or possibly 
IPE). A recent article by Thomas Clancy makes the neat and convincing 
suggestionthatlREcouldbeinterpretedas ire, Old Irish/Gaelic for 'inthereign 
of (Clancy 1993,345-53). AsClancypomtsout,rnkedGaehc/Latmmscriptions 
are known from Ireland although admittedly i re is not elsewhere recorded in 
any inscription. 

Another rmth-century inscribed cross-slab, that from Brechin, has no 
abstract sculpture or symbols but a complex set of Christian figure carving. The 
text, with its abbreviations expanded, reads SANCTA MARIA MATER 
CHRISTI, 'St Mary, mother of Christ'. This is not a memorial text but a 
descriptive one, describing the central carved figure. An example of an 
inscription on a Class 1 symbol stone is that from Brandsbutt, Inverurie, with 
the symbols ofthe serpent, the double bent rod and the crescent. Theoghamtext 
reads TRATADDOARENS- but, as is usual with Pictish ogham inscriptions, it 
has not been interpreted. 

An example of an inscription on a non-stone artifact is the eighth-century St 
Ninian's Isle chape, probably from a sword scabbard (Fig. 2.5). It is inscribed 
on both sides with texts that have been variously interpreted. One possibility is 
that they are to be read together as IN NOMINE DEUS RESAD FiL[I] 
SPIRITUS SANCTI O. 'ha the name of God, ofthe Son, ofthe Holy Spirit. 
Resad' . The final letter O is presumably an error and RESAD a personal name, 

S Britain 

Fig. 2.5 Chape from St Ninian's Isle, now in the 
National Museum of Antiquities of Scotland 
(photograph National Museum of Antiquities of 

probably that of the owner of the 

ThePicts then inscribed texts on 
a wider variety of objects than did 
the south-west Britons. Another 
point of comparison between the 
inscriptions of south-west Britain 
and those of Pictland lends further 
support to the view that the Picts 
had a firmer grasp of literacy than 
did the south-west Britons. In 
Pictland, the two scripts, ogham 
and roman, generally occur on 
separate monuments and are used 
for different languages. Texts in 
Latin were inscribedinromanscript, 
whether capital or insular, while 
texts in Pictish were inscribed in 
ogham. There are no certain 
exceptions to this rather rigid rule. 
It is true that the early Irish words 

cross 'cross' and magg- 'son' occur in some ogham texts but they occur as 
single words in the middle of Pictish sentences, presumably because they had 
been borrowed into the Pictish language. If IRE on the St Vigeans stone is 
interpreted as Gaelic ire, this may also indicate Old Gaelic/Irish borrowrng mto 
the language. In general, however, texts in Pictish occur in ogham scnpt, those 
in Latin in roman script. The existence of this pattern and the strictness with 
which it seems to have been applied suggest a level of literate awareness which 
is not demonstrable in south-west Britain. 

In the south-west there are no stones containing only ogham texts and no 
pattern emerges from the bi-literal texts. The Lewannick stone, for example, 
has a Latin text in roman script with only the Irish personal name repeated in 
ogham. The Fardel stone contains four different texts, two in each script, where 
two are single names and two may be either names or be in Irish. A correlation 
between language and script similar to that in Pictland is not found m the south- 

99 Bmmvn- 

A further point of comparison concerns the use of Latin. There is evidence 
of a greater use of Latin in south-west Britain than in Pictland, in that it is the 
usual language of the inscribed texts whereas in Pictland the usual language is 
Pictish. This may be associated with Bede's well-known account of King 
Nechtan's reply to Ceolfrith of Jarrow that he and his people would follow 
Roman custom in so far as their remoteness from the Roman language and 
nation would allow it (Historia Ecclesiastica V. 2 1). As noted above, the Latin 
used in the south-west is an irregular, if not a sub-standard, variety of the 
language. The same seems to apply to the Latin used in the Pictish inscriptions. 
There are spellings like [E]TT for et and FrL[I] for filii and some of the 
confusion of the St Ninian' s Isle text may well be due to linguistic errors in the 

Although the two areas can be seen to be in some ways comparable in their 
production of inscriptions, it can be argued that literacy among the Picts was 
of a higher standard and more widespread than literacy among the south-west 
Britons. A question that we should ask of both communities concerns the 
audience that was anticipated for their inscriptions. From our literate twentieth- 
century perspective it would seem a waste of time, money and effort to produce 
an inscription that no-one could read. Yet it is straining credulity to imagine that 
most people in, say, the ninth or tenth century could walk up to a carved cross 
or cross-slab like Sancreed in Cornwall or Brechin in Pictland, read the text and 
understand it. Still less is this likely to be true of ogham inscriptions. Who then 
formed the intended audience for these inscriptions? 

There would probably have been the occasional literate person, no doubt a 
priest or monk, who would perhaps have read aloud an inscribed text to other 
people. Some of the memorial stones, both those in ogham and those in roman 
script, could have been intended for that sort of audience in both areas. It may 
also have been that the inscribed texts were intended both for a human and for 
a divine audience. In view of the association between literacy and the church 
at this period, it would have been perfectly reasonable to assume that God and 
the saints were literate. If the saints in heaven knew someone's name then they 
could intercede for their soul. Of course it cannot be proved that the Picts or the 
south-west Britons argued in this way but we can speculate that they might have 
done so. 

S J Brvtatn, 

A further possibility is that inscriptions served a symbolic function. There 
nulwi hL ten people who, although illiterate, were able to recogmse 
Literacy was associated with the church and, we presume 
2£ XrTSgh status in the conrmunity. Perhaps it was considered that an 
£££ -ised a monument in status. The Brechin J^J^ 
above is inscribed with a text labelling the carvedfigure of the Blessed Virgm 
It SSy with such a well-known figure, that the carving was labelled to 
SXESScrian. Perhaps in this case the text was inscnbed as a symbol to 
^ate the importance of the cross-slab, or of the subject matter, or of the 
commissioner of the monument, or of more than one of these. 

We can conclude that in south-west Britain in the early Christian period 
t h e rwTaLlytenuousholdonl^^^ 

were it not for the inscribed stones we would have concluded that Dunmoma 

that they were produced over some five or six centunes. Moreover * .re is no 
surviving range of inscribed monuments, only stones, and mos of them are 
X crude pillar-stones. By contrast, literacy seems to have .played ^a more 

fwS r" artifacts. The inscriptions also exhibit a greater conhol of 

capital and cursive) for Latin texts. 

We are not really accustomed to extolling the Picts for then high level of 
lZly lT V 2I V s we should be doing just this. Some ten years ago I 


writing tradition, and that this was likely to have been a **^P^£ 
Ten though such manuscripts have by and large not survived Okasha 1985 

fof ? " dS— rP^ogham texts 
£££ SSJ— d the language and/or the speUing eonventtons 
nor are we certain of the meanings of the symbols on the symbol stones. 
How "e7h rs surely unjustifiably arrogant of us to argue that because we are 
fncITetent, the Picts were illiterate. We should perhaps grve greater 

3? Britam 

consideration to the idea of the Picts as a nation where literacy played a 
significant role in society. 

Dept. of English 
University College, Cork 


erences • 

Allen, J.R. and Anderson, J. 1903, TheEarly Christian Monuments of Scotland. 

Clancy, T.0. 1993, 'The Drosten Stone: a new reading',^,? 123 (1993), 345- 

Dark, K. 1992, 'Epigraphic, art-historical, and historical approaches to the 
chronology of Class I inscribed stones', in Edwards, N. and Lane, A. eds., The 
Early Church in Wales and the West, Oxford, 51-61. 

Nash- Williams, V.E. 1950, The Early Christian Monuments of Wales. Cardiff. 
Okasha, E. 1985, 'The Non-Ogam Inscriptions of Pictland', Cambridge 
Medieval Celtic Studies, 9 (1 985) 43-69. 

Okasha, E. 1993, Corpus of Early Christian Inscribed Stones of South-west 
Britain. Leicester. 

Thomas, A.C. 1994, And Shall These Mute Stones Speak? Post-Roman 
Inscriptions in Western Britain. Cardiff. 

Appendix - 

The texts of the inscriptions are given below, in the order in which they are 
discussed. The texts are transliterated with word-division spaces added, 
abbreviations expanded and incidental marks omitted. The following system of 
transliteration is used: 



5? Britain 

A indicates a legible letter A; [A] indicates a damaged letter, probably A; [.] 
indicates one lost letter, tbe number varying according to the number of dots; 
- indicates complete loss of text. A possible translation of each text is given 
inside inverted commas with damaged letters in square brackets and words 
added in round brackets. 

Lanherne, text 1: RUNHOL 'Runhol' 

text 2: BREID [ET] [I]MAH 'Breid [and] [I]mah' 

ordered this altar for his own soul' 

Welltown: [VA]ILAT[H]I [F]ILI VROCH[ANI] '(the stone) of [Va]ilat[h]us, 
son of Vroch[anus]' 

Southill: CVMREGNI FILI MAVCI '(the stone) of Cumregnus, son of 

St Just: SELVS IC IACIT 'Selus lies here' 

Lewannick: [HIC] IACIT VLC[A]GNI, '[here] lies (the body) of Ulc[a]gnus' 

Nanscow: VLCAGNI FI[LI] SEVERI, '(the stone) of Ulcagnus, son of 

Fardel, text 1: FANONI MAQVIRrNI '(the stone) of Fanonus, (son) of 

text 2: MAQIQICI '(the stone) of Maqiqicus' 
text 3: SAFFAQQUCI '(the stone) of Saffaqqucus' 
text 4: G[A]G[R]A[NV]I '(the stone) of G[a]g[r]a[nv]us' 

Lanteglos: AELSEL[D] 7 GENE [RED] [W]0[H]TE PYS[N]E [S]YB[STEL] 
FOR AELWYNEYS S[0]UL 7 [F]OR HEY[SEL] 'Aelsel[d] and 
Gene[red] made this ?family place (or place of peace) for Aelwyn's 
soul and for ?themselves (or Hey[sel])' 

St Vigeans: DROSTEN FREUORET [E]TTFORCUS 'Drosteninthereignof 
Uoret and Forcus' 

3 i Brltai 

Brechin: SANCTA MARIA MATER CHRISTI 'St Mary, mother of Christ' 
Brandsbutt: IRATADDOARENS- (uninterpreted) 

the name of God, of the Son, of the Holy Spirit. Resad ' 

95 Britatn 

36 Bi 


Prom Birsay to Tintagel : A Personal View 

Ckrlstopker D Morris 


As an archaeologist concerned with the archaeology of the First Millenium AD, 
ithas always irritated me that the term 'DarkAge' has survived. Itis itself arelic 
of a text-dominated approach to the latter part of this period, which has in the 
past emphasised either the absence of written sources or the confused nature of 
them, such that one might have described the period as much in terms 'text- 
confused' as 'text-aided'! However, much light has now been shed not only by 
re-examination of the texts and other text-related disciplines (such as place- 
names), but most particularly by the disciplines concerned with material 
culture (archaeology and art-history). The period concernedhas been illumined 
for some decades now by high quality scholarship on all fronts and deserves a 
more appropriate description. Terms such as 'Early Christian' and 'Early 
Historic' have been proposed in the past by notable scholars such as Charles 
Thomas and Leslie Alcock and certainly their currency in Ireland and Scotland 
respectively has been fairly widespread. However, neither term is entirely 
satisfactory as they have some 'value-laden' element embedded in them which 
tends to 'skew' the perception of the period, be it the dominance of the Church 
or the primacy of historical texts. It would seem far preferable to revert to a term 
whichissimplyperiod-descriptiveandofwidespreadcurrency-.Early Medieval. 

Two of the places most frequently referred to in relation to the archaeology 
of the Early Medieval period in northern and western Britain are Birsay in 
Orkney and Tintagel in Cornwall (Fig. 1 ). It has been my privilege to undertake 
archaeological work in both areas in the past two decades. I have indeed gone 
from Birsay to Tintagel and the work that I shall describe here looks at both my 
older work at Birsay, now proceeding towards publication, and the more recent 
surveys and excavations at Tintagel, which are only at an interim stage. This is 
very much a "personal view" on aspects of two excavation campaigns over a 
number of years in two widely separated parts of Britain, although broadly 
contemporary. It will be concerned with three aspects: 
1) the evidence, as we now understand it, for Birsay in the pre- Viking period, 
when I think we can reasonably claim it to be a power-centre of some 

37 Hm 

2) a re-consideration of the previous proposition that it should be seen in the 
context of a 'Celtic monastery' model; 

and 3) a brief review of the evidence from Tintagel, which has also been seen 
in such a context, and suggestions as to how we might now interpret it. 

Birsay Bay 

The Bay of Birsay is a marked indentation on the N W coast of Mainland 
Orkney (Fig. 3 .1). It is, in fact, two bays, divided by a small promontory of land 
called the Point of Snusan or Snushan. The northernmost is the larger, bounded 
on theNbythePointofBuckquoy,whichwas originally attachedto the Brough 
of Birsay, now separated by the 238m wide Brough Sound. This tidal island, 
projecting out into the Atlantic, is connected at low tide by a modern concrete 
track across the natural causeway of exposed rocks. 

Fig. 3,1, Location of Birsay and sites around Birsay Bay (L McEwan 
after N Emery) 



It is clear, as several writers have emphasised (Marwick 1952, 130-1; 
Crawford, B 1983, 116-7), that the name 'Birsay' is a contraction of the form 
Byrgisherad, as found in Orkneyinga Saga (although the form Byrgisey is also 
found). The second element, heral refers to the administrative district, later 
parish, of Harray, while the first is derived from byrgi, the i-umlaut form of ON 
borg, a fortress or stronghold. The term in the Faroe Islands ".. .is almost always 
used ofnarrownecks or steep cliffs orofenclosedplaces for sheep" (MacGregor, 
in Morris 1989 Chap 1.4), and this would be an entirely appropriate description 
for the first element of Byrgisherad. Such a narrow heck of land might well 
have originally existed - rather as had clearly existed at the Brough of Deerness 
(Morris with Emery 1986, 309-10) -and perhaps represented on the 17th and 
18th Century drawings of the Earl's Palace and Birsay (RCAHMS 1940, n, 
Figs 68 & 69, Pis 5 & 6 after 12). 

Historical evidence indicates that in the past Birsay was a centre of both 
political and ecclesiastical power in Orkney second only to Kirkwall in the later 
Norse, Mediaeval and early Modern periods, but forerunner to that place in the 
Viking period. However, it is only with the accounts in the Orkneyinga Saga 
(OS), written c. 1 1 92-1206, of the exploits of the later Viking Earls of Orkney 
that Birsay itself enters the historical record. Two entries concerning Earl 
Thorfinninc. 1048 andc. 1065 (Taylor 1938 (ed), 188-189, 368,n3) establish 
quite clearly that both the Earl's seat and the first Cathedral were in Birsay (OS 
chaps 31 and 32: Palsson and Edwards 1978, 71). There has been much 
discussion as to the significance of the Saga entries for location and identification 
of these structures, and this has been well-rehearsed elsewhere (e.g. RCAHMS 
1946, n, nos 1 & 6: 1-5 & 7; Radford 1959; 1962a & b; 1983; Cruden 1958; 
1965; Lamb 1974; 1983; Cant 1983, 8-9). Birsay became the political and 
ecclesiastical power-centre of an Earldom, which extended N to Shetland, and 
S at the very least to Caithness, and at times probably far wider. Later, Birsay 
became an area particularly associated with Orkney's own martyr-saint, 
Magnus. The sequence of events concerned with his death, burial in the minster 
at Birsay and the subsequent miracles reported at his grave are re-told in some 
detail by the Saga writer, cuhninating in the translation of his relics to Kirkwall 
(OS chaps 52, 56 and 57: Palsson and Edwards 1968, 88-9 and 94-9). Norse 
Christianity clearly focussed upon Birsay, but once Magnus's bones were 
transferred to Kirkwall, and the Cathedral was built there, naturally the focus 
of secular and ecclesiastical power was shifted away from Birsay. 


0*Wy uv^Pve-Norse pe^A: K^al con^ai^ 


view, the picture in Birsay ts ^-totono teto ^ ^ ^ 

references whatsoever. However, **™™J^\ bave a certain amount 
of historical or quasi^^ 

native people m"theNo rt he m Is^ described the islands as 
A late Scandinavjan sovnc $ to fi^J ^ ^ , 33 0-l ; 

having been inhabited by Picts anor p v whom appgar tQ 

Crawford 1987, 3 & 56; W^^J ^ such as Papa 

^W^^rfJ^S^ Papdale hi Orkney 

Stour and Papil in Shet and and Papa W^ay « * 

(MacDonald 1977; Crawford 1987, 165 /, mo 

« 800 AD there was no doubt that Picts "occupied the 
To Nennius, writing c. 800AD,tnerewa an d aless 

W-d. which we cal ^g^^SSS^W^ 
reliable source, Claudian (chap 32^quo ed ^ a g ofthe5ernC/irom cZ e 

trans) 1978, chaps 14 and 19). 

their heterogeneity at the time when th J^\™* ld be seen as a 
AD onwards), and it would now seem d» that ^^ ey ^ formercoUeague) 

Professor JohnMann,has compared ^"^^ fac S e 0 f Roman 
^^=a;l^M^soseeBree Z el99 4 ). 

Bede's statement that 

the speakers of four different ta«W*Bn ^ ™ ave 
English..." (Hiftoria Ecclesiastica (HE) m, 6. coi D 
Mynors (ed & trans) 1969, 230-1) 

*9 Britain, 

indicates that by the seventh century the Picts were recognised to be one of the 
four great peoples of North Britain. 

Even so, indications that the Orkney Picts may have acted as an independent 
unit (or as part of a "northern Pictish" grouping) seems clear from accounts 
of the putting down of a rebellion there in 682 AD by Bridei mac Bili 
(Tigernach Annals & Annals of Ulster. Anderson (trans) 1922, 1, 191), and of 
the reference to an Orcadian regulus inc. 565 AD (Adamnan, II, 42: Anderson 
and Anderson (ed& trans) 1961, 440-3) inrelation to a. visit by St Colurnba of 
Iona to the court of Bridei mac Maelchon near Inverness and in the context of 
the voyage of Cormac, a follower, who 

"...attempted for the second time to seek a desert place in the 
ocean.. .[and] The saint ...foreknew in the spirit that after some 
months this Cormac would come to the Orcades. And it did 
afterwards so happen" (Anderson and Anderson (ed & trans) 1961, 
chap II, 42, 440-3). 

In addition to the "mission" of Cormac, about twenty years later, theAnnak 
of Ulster record one or two expeditions (there may be a scribal repetition) of 
King Aedan mac Gabrain to Orkney from Dalriada (c. 580 AD: Anderson 
(trans) 1922, i, 86). Certainly, with this background, there seems no reason to 
doubt the likely influence from the Columban church in the area of Northern 

However, there were other influences, the most obvious being that from 
Northumbria. After the Battle of Nechtanesmere on 20th May 685, the 
Northumbrians accepted a border - metaphorically at least - at the Firth of Forth 

"The Picts now have a treaty of peace with the English and rej oice 
to share in the catholic peace and truth of the church universal..." 
(HE v, 23: Colgrave and Mynors (ed & trans) 1969, 560-1) 

The apparent wish to "share in the catholic peace and truth" indicates a shift 
in ecclesiastical alignment on the part of the Picts towards the practice and 
outlook of the Northumbrian Church in the post-Synod of Whitby era. It is, 
therefore, unsurprising that in 7 1 0 AD Nechtan, King of the Picts, should have 

*J Brvtafn 

appealed to Ceolfnd, Abbot of Wearmout^^^ 
subjects of the date of Easter and the form of tonsure. But, rn addition. 

"He askedforbuilderstobesenttobuildachurchofstoneintheir 

(ed& trans) 1969, 532-5) 

This Northumbrian influence can only have intensified in the period after 
716 when the expulsion of tefamilia of lona from Pictish territory took 

Althou* there is no necessary connection with Orkney in the legend of a 

hnih fLeeend of StBoniface in Skene 1 867 ; Skene 1 876, i, 277, u, 229 , apnon 
SSSSSSd as a distinct possibility that, if there is any historical basvs 

foXleg Z,tor^^^i^^*»^$^ 
he rel ability and significance of both church dedications and Pa^ularly 


of cnurlhes relate in part to this phase of activity. The smgle Boniface 

taken to be of this period in Orkney, and may also link up with the St Tredweh 
(Triduana) dedication, also on Papa Westray (Thomson 1987, 9-10, Lamb 
1995). William Thomson has suggested that: 

"Atthe very least, it demonstrated the presence of a cult which had 
its base inNorthumbrian territory, and suggests links between the 
OrkneyPapaeandaNorthumbrianmissiontoPictland (Thomson 

1987, 10) 

isuaralleledby a Carolingian form of ecclesiastical organisation (Lamb 1993). 
So Sesfmore credit contextforthe occurrence o^ 

the suggestion of an eremitical movement to — s— 
whichmanifestlyinseveral cases theywerenot(MacDonaldl977,Lowel98«, 

Appendix 3; Thomson 1987, 40; Lamb 1993). 


this pSoTwe g may be sure that pre-Norse archaeological material found 

*J Battel, 

during the course of archaeological investigation is most likely to relate to a 
context of the historical Picts, and in particular to an Orcadian sub-group of the 
Picts. This Pictish sub-group, however, is also likely to have had connections 
with both the Dalriadic and Northumbrian political and, more particularly, 
ecclesiastical and cultural spheres of influence. 

Birsay in, the Pictisk period : the archaeological evidence 

In the absence of specific documentary evidence for the status of Birsay in the 
pre-Norse period, we may now gather together the results from the various 
more recent archaeological investigations for an up-to-date characterisation 
of the material evidence for the Pictish period in Birsay Bay. 

■phases X-X 

Fig. 3.2a. Buckquoy: plan ofPhasesI-II(IGScott, 
Crown copyright) 

Settlements and Houses 

Naturally, pride of place must go to 
the breakthrough in understanding 
of Pictish settlement types. 
Although her interpretation of later 
sequences has been somewhat 
controversial, the Pictish phases at 
the site called "Buckquoy" 
excavated by Dr Anna Ritchie have 
provided clear examples of 
distinctive buildings in each phase 
of activity. The early Pictish 
houses at Buckquoy were both of 
a cellular form, one being trefoil 
and the other withfive cells (Ritchie 
1977, 176, fig 2), (Fig. 3.2a) 
although the exterior forms were 
quite likely to be oval (Ritchie 1989, 
47). The later Pictish farmstead on 
the Buckquoy site was far more 
elaborate, even prompting the 
description "almost anthro- 
pomorphic" (Ritchie 1974, 27; 

« Britain 

Ritchie 1977, 183; Ritchie 1983, 
56). (Fig. 3.2b). This form is 
certainly a sophisticated one, in 
contrast to the "cellular" forms, 

and was immediately related to 
similar buildings found by Dr 
IainCrawfordattheUdal, North 

Uist as well as other, less well- 
recorded or understood sites in 
the north (Ritchie 1974, 26-9; 
see Crawford, I A, 1974). 

When the site of Buckquoy 
was seen as anisolated domestic 

site in the area of Birsay, 
questions obviously arose as to 

of Birsay (Ritchie 1983, 54) and 
the excavator postulated that "it 
is possible that Buckquoy 
functioned as the home farm for 
the community living on the 

Fig . 3.2b. Buckauoy: plan ofPkasesII-VI (I G Scott, 
Crown copyright) 

5^ o** 'rr w ^7-tz:rrrf 

an isolated site between the two well toownjo . ^ples 

which survived only in fragmentary form. 

Morrisl989,Chap S 6&73-7^ 

figure-of-eight shape 3 ^"^^ from a flagged threshold to 
was in the S wall of the W room, ossiblehear that1heE end of 

someintemal flagging. There °Cm B and a stone - lined 
Room A, but there & ^.^^ed what is interpreted as an oven. 

the later phases of collapse 

corbel -fashion, and in one 


Fig. 3.3. Red Craig (Area 3): photograph of main structure 
from W (C D Morris, Crown Copyright) 

area of the site - to the E - 
there was some evidence for 
the presence of a surrounding 
annulus (Area 4, Phase A). It 
would thus seem , that a 
hollow had been delib erately 
created in the natural subsoil, 
which was lined with stones. 
In this case, a substantial 
horizontally-coursed wall 
was created on the inside, 
and, despite later destruction, 
it would seem that there was 
then an encircling bank of 
clay with further walling 
forming arim or edge to this. 
Final publication has stressed 
that, while its internal form 
is distinctive (its nearest 
parallel in shape, and possibly construction, being the double clochan at Reask 
in Co Kerry), this is likely to be 
a variation upon a common 
theme, with a less distinguished 
exterior appearance as at 
Buckquoy (Morris 1989, chap 6 
esp 171-2 & chap 10, 287). 

The remnants of what is 
assumed to be a second building 
of a form perhaps similar to the 
circular cell at the "head" end of 
the late Pictish building at 
Buckquoy was also uncovered 
on this site (see Morris 1989, 
chap 6, esp 189 & chap 10, 283- 
5) (Fig. 3.4). It consisted of a 
"scoop" into the natural subsoil 
which had been lined with stones 

Fig. 3.4a. Red Craig (Area 5): photograph of remains 
of second structure from N (C D Morris, Crown 


to create a circular room; this 
W as divided off from what 
presumably was the mam 
room by a partition 
holes originally filled wvth 
vertical slabs. The 
circular room, with vertical 
slabs, is reminiscent of the 
Late Pictish building at 
Buckquoy (Ritchie 1977). 

It does not seem 
unreasonable to interpret the 

two buildings together as a 
small farming-unit, with 
perhapsayard to be implied 
to the S of the figure-of-eigM 

meartefacmal materia^ 

Edification to, the figure-of-eight ^™ rigW flagstones. Perhaps 
Room A were replaced by ^^^^JJoa of what appears to 

m0 re significant in ^^^^^ AM ^ %VU ^ 
havebeenasolidstonepartitionwallbew ^ f . 

to both N and S. It may evidence to support the idea 

thatitmayhavebeenat Uy ^sh" artefacts; 

lt 1S to be noted that there ^^^^udcquoycolle^m 
mdeedtheonlyartefactfromRedCraigtha parahe toargue&0 m 

fa stone gaming-board from Phase fe ^ 

the ateence of clearly identifiable Norse c ^ Qr bone 

assemblage from the early ^^^iarg^hDwever^lcart 
Negative evidence is always difficult tc ^^^s^orechar.ctensttcaUy 

Viking-age artefacts were present abov 


Fig. S, b . B* Cra ig (Area ff^o^CrZ 
Jhin second structure from W (C D M 


shapedbuilding,butnowdestroyedby erosion. 

for the later phases are AD 600-910 (Phase C: occupation) and AD 875-1055 
(Phase E: disuse). This would then imply that the construction and occupation 
of the structure should be dated to the Late Pictish and Early Viking periods, 
and its disuse clearly to the Viking period proper. 

The nature and parallels of the structural forms are crucial to identification 
of cultural context. It is now a commonplace that Pictish-period house forms 
are cellular in form, be they "radial" or "axial" (Hunter 1986, 26) and that they 
contrast markedly with the buildings of the Scandinavian incomers (Crawford 
1987, 140-6, esp Fig 46). Despite the likely exterior form as an oval, it is quite 
evidently the case that the Red Craig building is related to this native "cellular" 
tradition, a tradition that can also be seen in a group of such buildings at the 
Point of Buckquoy, originally examined by F T Wainwright (see Morris 1989, 
Chap 4). 

Recent work by Dr John Hunter on the Brough of Birsay produced a series 
of primary features interpreted as structural remnants from his "Phase 1". 
Although all these features on Sites VII, VIII andIX imply "cellular" structures, 
only sufficient remained on Site VII with structures 19 & 20 for convincing 
reconstruction (Hunter 1986, Chap 2, esp 39-45 & 61-4) (Figs. 3.5 and 3.6). 
Further structural remains excavated by Morris on the Brough of Birsay, 
identified by the presence of primary features, such as post-holes below Norse 
buildings on Site IV South, or gulleys around buildings whose walls had been 
removed on Area I, appear to date from the pre-Viking period, from the 
presence of an ogam stone (see below) and insular metalwork (Morris 1 98 1 & 
1982; Hunter 1983, 155-6; Morris forthcoming a). Radiocarbon dates are 
consistent with this, giving a range at 2-sigma level from early seventh to late 
eighth century. 

It is notable that the two date ranges accepted above for Phases A and C at 
Red Craig are consistent with the overall range obtained from Sites VII and DC 
on the Brough of Birsay (from AD 640 +/- 60 to AD 828 +/- 70). No 
radiocarbon dates were utilised at Buckquoy; but the Late Pictish period dating 
derived from artefacts and retrospective chronology relative to the Norse 
phases (Ritchie 1977, 192) is internally consistent and does not conflict with 
other historical or cultural considerations. Overall, the overwhehriingprobability 
is that all three sites are contemporary and reflect the same Late Pictish period 
cultural milieu. 

*J Britain 

1984, 12-15 & 36; Gelling 1985), and examples at the Wag of Forse and 
Nybster in Caithness and Gurness in Orkney are among comparative structures 
brought forward by Ritchie(1974,26-7&32)andAlcock(1984, 12& 18-19). 
More recently, Beverley Smith has drawn attention, among the variety of forms 
at the Howe, to a small rectangular building (Smith 1990, Illus 4.2e & 37). 

Fig . 3.5. W ofBirsay (SUeVQ^ 1 «*« '* " & " 9 

R Hunter, Crown copyright) 

2(KeMduel983,63)b««m^^^^ ; afield: to 

nave continued to be excavated ^^^L * al 1984; Ned 1985; 
Howe, near Stromness (Hedges and Be l 198U, ^ Broch rf 

SU, 1985, 201-2; Smith 1990 £P ^i atd 132 Ritchie 1974, 
Gurness (RCAHMS 1946.D no263 75-9 espfigsU ^ ^ rf 

26, fig l; Hedges 1987, II, 65-71 198^196 & 201); Yarrows andthe 

Ed^C^^^^S^i! ^ & 32); Carlungxe, Angus 

and Armit 1990, 91-2 & 98-106). 

Fig. 3.5. Brough ofBirsay (Site VII): plan of Phase 1 structures 19 & 20 (J R Hunter, Crown 

When Alcock reviewed the traditions of building of the period, he 
suggested that there were two traditions: "circular" and "axial" (Alcock 1 984, 
As Ritchie has implied, this is likely to be an over-simplified division: 

"It seems likely that these various house-types represent inter- 
related architectural development rather than separate traditions, 


related architectural development rather than separate traditions, 
and the hnpression given by the existing dating evidence is that the 
development is linear and sequential from wheelhouses onwards" 
(Ritchie 1985, 201) 

4§ Bxrvfil 



* Britain 

Certainly, although the Buckquo y-rela* ^^^^ 
readilyrecognised^itisclearthatthere ^^^*S^ tdflVaIlt tD at 

leastadiscussionofthe>oto-Picts C^"^T ^ 19 80; 
as W ell as cucular stores ^^^^^^ 

"...the nucleus of the later Iron Age seu ' " atPool, Sanday 

(Hunter 1990, 181). In between the *vo «amm befce ^ 
'blocks', with oval and sub-rectangu ar rooms f ™f^ 1990 183 ^ & 
(Ritchie 1989, 48; Smith 1990, Illus AM. e & f & 3 ' 
esp Illus 10.6). The structures it must be 

Bay (Hedges 1983, 78-81) fit ^^^Lfin^ 
rememberedthatsomeofthebuJdmgs ' s ™P^ an dwelliBgs (Ritchie 
ma y have been workshops and storehouses ^^^^^Lceto 
198 9, 48-9). The complexity of the srtuatronts best VP^*^ (Smith 

(ed)1994,Chap5).However,it 1 swoAno to g(a« 
IpartftomSaevarHowctheBusaysttesare de^daed « 
contrast to the 'clustered' complexes at Pool ^rl9^ * 
(Smith 1990; Smith (ed)1994) and, by rmphcation, Cmp and Locn 
(Armit 1990; Harding and Armit 1990). 

Symbol Stones and Graves 

The symbol -"^Sg^ 

1965,25) N<m ^^^^j 1994,19; Patchie,A, 
the case (Curie 1982, 91-2, Kitcnie, . • f t since itwas 


ibnndmptac^^ J nt work , for 

to some other g*^" 00 ^^.^^^ 1980" Close-Brooks 
instance at Dunrobin ^^^^^^^^o^ch^A 
19 8 4),andWatenan,Cai« 

that there is "...more and more c i rcu ™™ , Ashmore 1980, 

stones were gravestones" (Close-Brooks 1984, 107, also 


5 i> Britain- 

It now seems clear that there was another Pictish stone also from Birsay: a 
decorated stone covered one of the short cists in the cemetery overlying the 
ruined broch mentioned above at Oxtro, with "the figure of an eagle... boldly 
cut" (Petrie 1873, 76-8). Although now lost, this seems best explained as a 
Pictish symbol-stone (Wainwright 1962, 93-4; Ritchie, JN G 1969; Ritchie, A 
1985, 188 & 190-1; Morris 1989,24). Presumably it was re-used at Oxtro, but 
would have been a separate monument originally. 

The symbol stone on the Brough of Birsay was associated with the lower 
stratum of burials in a cemetery seen by earlier excavators as pre-dating not 
merely the upper stratum of Norse Christian graves but the Viking period as a 
whole (Cruden 1958, 160; 1965, 23-5; Radford 1959, 17-18; 1962, 11; Curie 
1982, 13). The graves took the form of long-cist burials with horizontal slabs 
and, in many cases, small head- or foot-stones. These graves were also seen as 
associated with structural remains below both the later chapel and the Norse 
churchyard enclosure, and the 'Celtic' nature of the remains appeared to be 
reinforced by the presence of what was taken to be a leacht (Cruden 1965, 24- 
5; Radford 1959, 17-18; 1962a, 11 & Plate II). Although a number of these 
assertions might now be questioned, there can be little doubt that there was a 
major cemetery of the Pictish period located on the Brough of Birsay. 

The edge of part of a Late Iron Age and Pictish cemetery was excavated 
under rescue excavation conditions in Areas 1 and 2 on the Brough Road South 
of Red Craig (see Morris 1989, Chaps 5; 7.1-7.2; 7.6 & 9). On top of the 
naturally-deposited sand at the base of the excavations were set two stone cairns 
over human burials. Cairn 2, the more completely preserved example, 
demonstrated the construction method. A stone cist (in this case incomplete) 
had been set in the underlying sand, with an extended inhumation on a NNW- 
SSE orientation, placed on its side (Figs. 3.7a & 3.7b).Over this was heaped a 
layer of clean, barren sand, and then a mound of irregular stones and sand, 
which were faced with an outer kerbing of horizontally-coursed sandstone 
slabs, seven or eight courses high. No grave-goods were found in association 
with this burial, which was of a mature, possibly elderly, male. Cairn 1, much 
less well-preserved, but apparently of similar form, covered a cist (of much 
better construction than that under Cairn 2) in which two burials had been 
placed. The lower was probably an adult male, the upper an individual of about 
18 years of age, whose sex is unclear. Only the lower halves of the bodies 
remained, due to erosion, but it was nevertheless likely that the lower torso had 

B Britain. 

found with either skeleton. 

Crown copyright) 

the radiocarbon dMB—ons. From the « the p ^ 
were elearly to be associated with the earlier P >* sWK0 230 . 5V0; AD 

245-585 (Morns 1989, 288 ana/**;, * seems 

RomanlronAgeandearlierpartoftheP ^^0^^ in cultural terras^ 

and to describe these as "Pictish burials The rec ' J ^ 
bones, some of which at least are earhe ]^^Z^?^Bmd 
cairns, justify the description of this overall gwen a terminus 
C above the burials produced no diagnostic artefacts, but ax :g ^ 

52 Britain 

(Ritchie 1974, 31-2; Ashmore 1980; Close-Brooks 1984). Here the Pictish 
tradition appears to have been that of long-cist burial below mounds surmounted 
by well-constructed kerbed cairns, whether rectangular or circular. It has been 
suggested that the circular form (here and as already identified at Keiss, 
Ackergill and Watenan in Caithness, Garbeg and others) may be earlier than the 
rectangular (such as at Dunrobin and Sandwick) and that the latter may have 
been in usage amongst the Picts when they came into contact with the Norse - 
perhaps prompting "borrowing" of the form (Close-Brooks 1984). The long- 
cist form is ubiquitous in relation to these cairns, and the mound of sterile sand 
a common feature. 

The long-cist form of burial is also a contemporary tradition that was 
followed in cemeteries where no cairns were constructed. In the light of the 
association of isolated long-cist burials with cairns at some of the sites now 
clearly identified as Pictish period cemeteries, it seems reasonable to interpret 
the feature from Cutting 3 on the Brough Road in similar terms, and to assume 
that the isolated human bones found through the early deposits of Area 1 may 
have come from similar burials. A similar isolated cist grave was excavated by 
Dr Ritchie at her Buckquoy site (Ritchie 1977, 183-4; 1983, 61-2), and it is 
notable that Pictish period slab-lined graves, dated by radiocarbon (although 
the details are not presently available), have recently been identified from 
Westness, Rousay (Kaland 1993, 312-4). 

Ogam Inscriptions 

So far, seven Orcadian Pictish period ogam inscriptions have been found, with 
four corning from the Birsay area (Ritchie 1983, 52-3; 1985, 191-2; Hunter 
1990, 185-6). Two recorded from the Brough of Birsay in the past have 
uncertain original locations (Radford 1959, 5; Radford 1962b, 174; Cruden 
1 965, 25; Ritchie 1985,1 92), but the location of another is certain, having been 
found during excavations directed by the writer on Area IV South (Morris 
1981, 36;Morris 1982a, Plate 41) (Fig. 3.8). The fourthis on the, bynowwell- 
known, sandstone spindle whorl from Dr Ritchie's site at Buckquoy (Ritchie 
1977, 181-2, 197,fig8no 84, 199,221). It is interesting that this particular one 
was on a portable object, and Ritchie wonders if it is to be explained in terms 
of talismanic or magical connotations (Ritchie 1985, 192). But it is worth 
noting - as Oliver Padel has done (referred to in Ritchie 1985, 192) - that the 
object upon which this inscription had been placed was indeed one of everyday 
use, for which he coined the phrase "chattel inscription". 

?? Britain 


of letters and therefore, as suggested by Jackson * , otflie p ic Ush 
,86), drey were conveying <^*J^S3S^i»-i^»-'. 

Orkney, the earlier discoveries "^"^f Th e recent 

^S;ttr^vS ( co»« a se^by ambble spread 
dated by radiocarbon to AD 560-769. 

have important implications which she has dvscnsse 


they are carved by different hands. They are described by Dr Forsyth as 
"informal graffiti", with perhaps fragments of personal names involved, of 
which Birsay 3 (from my own excavations) is the most intelligible, perhaps 
even referring to a name such as 'Mac-Onchon', as recorded in the Book of 
Leinster. Again, the indications are that the inscriptions are likely to be Irish in 
origin (Forsyth forthcoming b). 

3. A small pebble may also have ogam characters (Curie 1982, 120), but this 
has yet to be verified by Dr Forsyth. 

It is quite obvious that this re-evaluation is very important, implying as it 
does an Irish linguistic presence in the Birsay Bay area in the seventh and eighth 
centuries. It is also important in relation to the possible influence of the 
Columban church mentioned above, and the tradition of St Findan and Orkney 
(see Morris 1990, 7, with references). 


Another, apparently distinctive artefact type recognised in Birsay, is the 
bronze-bound iron bell: two have been found. The hand-bell is normally 
associated with an ecclesiastical context and seen as a reflection of the activities 
of the church in the 'Celtic' areas of Britain and Ireland (Anderson 1881, 167- 
215). More recent work by Cormac Bourke on Irish, Welsh and Scottish 
examples suggests that, when found outside Ireland, they reflect the influence 
ofthe Irish church (Bourke 1980; 1983). The nineteen from Scotland (of which 
fourteen are iron) perhaps then reflect Irish missionary activity mediated 
through Iona (Bourke 1983, 466). The example from Saevar Howe is quite 
well-known, having been found in the nineteenth century (Fairer 1862; see 
Batey and Morris 1983, 85 and 97-8) (Fig. 3.9). Now, however, we know that 
the context for the bell is no longer so clear as the nineteenth-century account 
and interpretation would seem to indicate. This implies connection with an 
Early Christian burial-ground, and deposition with the advent ofthe Vikings 
to the area (Anderson 1881, 170-1; also see Cruden 1965, 25). Hedges's re- 
excavation demonstrates quite clearly that the cemetery post-dates the Viking 
and Pictish buildings on a settlement-mound site (Hedges 1983, passim), 
although the precise context of the bell as found in the nineteenth century within 
this is unclear. It may either be associated with the later cemetery (Morris 1983, 
141) or be distinct from this cemetery and thus possibly be earlier (Hedges 

S Britain 

1983, 120-1). Some support for 
the later dating possibility might 
be given by the evidence of the 
of Birsay, apparently from a 
Lower Norse horizon (Curie 
1982, 50-2; Curie 1983,77-8). 
However, it should be noted that 
the context for this is in a hearth 
of a building of some 
sophistication (Room VI), dated 
by Radford and Cruden to the 
eleventh century (Radford 1959, 
19-20; Cruden 1958, 160-2; 
1965, 29-31). Since the context 
must surely be secondary, as with 
Saevar Howe, the bell could still 
be early and from the Pictish 

Fig. 3.9. Saevar Howe, Birsay: drawing of iron bell 
(H Jackson, Crown copyright) 

Other distinctive artefacts 

own site inBirsay atBuckquoy (Ritchie 1977, 182, fig 8, no /, w 
58-9; Ritchie 1985, 200). 

from Red Craig is likely to be Late Pictish/ barly V W 

286), whereas that for the stratr Bed ex^g eft mBoc kquo y * ^ 

associated with the f^f^ from the Mrddle Norse 

of whalehooe , « frond oo the o ^ ^ „ sterclx , s 

B i Britain, 

of 'brandubh, it is also essentially the same as Scandinavian hnefatafl (Sterckx 
1970; 1973a; 1973b). CouldithavebeenadoptedbymemcomingScandinavians 
from the native Picts, and thereafter spread throughout the Viking world? 

Evidence for fine metal-working and other crafts 

On the Brough of Birsay, the excavations below Room 5 in the areata the east 
of the church have demonstrated that the lowest phase, stratigraphically, 
contained clear evidence for metalworking (Hunter and Morris 1982). This 
can, as shown in the report, be dated to the Pictish period, on grounds of both 
radiocarbon dates and other associated artefacts. This then has provided a 
context for Mrs Curie to analyse the material collected previously from the site, 
and her Phase plans for the area to the east of the chapel (Curie 1 982, 1 5-17, 111 
5) demonstrates clear evidence for Pictish working "zones". Crucibles, slags 
and moulds from "Zone 5" were related to the evidence of moulds etc from 
previous excavations (Curie 1982, 26-44). One of the brooch-moulds from 
belowRoom5(no300: Curie 1982,26-7,111 13 & 14)canberelated specifically 
to the metalwork from StNinian's Isle - as can other moulds. Moulds for other 
copper-alloy objects, including multiple pin moulds (which match bone pins 
foundbelow Room 5) and ornamental pins, have been found, with a significant 
concentration around a small well. The overall impression is that "Birsay must 
have been an important centre of bronze-casting before the arrival of the 
Norsemen..." (Curie 1982, 26-7). Similarly, a noteworthy lead disc, which 
might have been used for cire perdue casting (Curie 1974; Curie 1982,48-9), 
crucibles and tuyeres reinforce this impression (also see Curie 1983, 71-7). 
From the recent excavations a fine gilded bronze mount, with clear parallels 
with Northumbrian material of the pre- Viking period, has been recovered 
(Morris 1982, Illus 42; Cronyn, Morris and Owen 1983). 

Although there is no debris from glass-making activities, an amount of 
evidence is present which might suggest small-scale craft activity using glass, 
for instance in bead-making or inlay-work in metalwork (Hunter 1982, 46-7): 
glass mounts and a waster, vessel-glass (from both earlier and more recent 
excavations) and a tessera are noteworthy in this context. 

Overall assessment 

As is clear from this review, there is a considerable amount of archaeological 
evidence to demonstrate the importance of the Birsay area in the Pictish period. 

I 7 Br-ltain, 

conceivably, power - as it was later to Dec faAtos been 


towidenthe^chaeolo^alc^fo *er ^ bulldings and internal 
perspective for the area ^J^; J^^p^talh-I^ 
occupation debris from Red Craig proviu a i on g with Hunter's 

Pictish settlement atBuckquoy. They ^^^^ 8 stI ^fi J n n s 
r ecentworkontheBroughofBirsay,theva^tyo^ 

farmingsettlementsatthistim^ .forth ^^J 0>e ^ evidence 
at the Point of Buckquoy, a further c. 00m away ^t ^ unpub i ish ed) 
from the Bay for Pictish period burials ^-^^S long cists from 
late Christian cemetery on the B rough. Th » 

chronological range. 

«•■*.« ^Pir>ti«h and Viking settlement in the Bayfhat 
TheclmgedoverallpxctureofPictishmd^g majQr 

necessarily derives from the w^^ the native Picts to the 
importance to the question of the mter ^^ 

incommgNorsemScotland,asBirsay^ ^ 155 . g8) . 

1983; 1985, 191-198; Morns 1985 21 ,6-22 ; ^f^ e may note tha t, on 
This is not the place in which to ^^^^^^ Earldom of 

Orkney may represent the takeover p hFirthofthe distinctive 

events to the south or west (i.e. as tbey rrnpmg v Y and 

Northumbrians or Dalriadic Scots), and ^J^J &s 
archaeological evidence is concerned n ^ ^~ ofessorLes He Alcock 
is now patently clear, they Kenneth Jackson (1955), 

(1984, 9-10), perhaps over-influenced by Wotessor 

s i Britain 

has contrasted the so-called "peripheral" Picts of Northern Scotland and the 
Northern Isles, with the so-called "heartland" Picts, so that the former supposedly 
had "a diluted Pictish or Proto-Pictish culture". However, as Dr Anna Ritchie 
has observed, Orkney was no more peripheral than Angus at times in the sixth 
or seventh centuries (1983; 1985), when the Moray Firth (i.e. the Inverness 
area) appears to have been the centre of the Pictish realm (Henderson 1971; 
1975). Indeed, some of the finest examples of the Pictish sculptured stone series 
are to be found in this area (Curie 1940; Henderson 1958; Close-Brooks 1989; 
Henderson 1990). The work of the last two decades in Orkney - andparticularly 
Birsay - have now amply demonstrated that conventional views of "centre" and 
"periphery" must now be re-assessed for the Pictish period in northern Britain. 
The "Northern Isles" of Shetland and Orkney were integral to the Pictish 
cultural, and presumably political, orbit and Birsay would appear to be a major 
power-centre within this. 

The Church and 'The Celtic Monastery' 

Some time ago, Dr Ralegh Radford emphasised the place of Orkney in the 
overall story of "The Celtic Monastery inBritain" (Radford 1962a). His wide- 
ranging discussion to ok in Ireland, Wales, Cornwall and Scotland and included 
both the Brough of Birsay and the Brough of Deerness, as examples of a 
monastic type to be seen most clearly at the site of Tintagel in Cornwall. This 
was elaborated upon in a further paper, in which Papil in Shetland was added 
to the probable list, and Eynhallow ("holy island") in Orkney was considered 
as "a small Norse monastery". Other sites were briefly considered in the context 
of "hermitages", for instance St Ninian's Isle, Shetland, or as "churches of 
lesser consideration", for instance St Tredwell's chapel onPapa Westray. With 
the latter, a connection with the Pictish area of activity was accepted, rather than 
the "Irish west" (Radford 1962b, 165-9). Essentially, Radford held to these 
views in his paper on "Birsay and the Spread of Christianity to the North", 
published in 1983, merely amending his interpretation of St Ninian's Isle, by 
analogy with the Brough of Birsay, from hermitage to monastery (Radford 
1983, 18-20). 

The site of the Brough of Birsay, now a tidal island, is one that lends itself 
to such an interpretation and there are a number of individual features which 
have been cited in support of this. Radford has considered one excavated 

9 B^tatn 

i Radford 1959 17-18; Radford 1962a, Plate II (b); 
feature to be a Zeacfe (Radford TO, w » emphasised its 

Radford 1962b, 168-9; ^^^^^kN^-dto 
positionwitrnnacenretery ^^ q %^ This cem etery has been 
earlier therefore deduced to be Picttsn or & 

small stone church at the centre of th gular enclosure (Radford 

as a curvilinear enclosure underlying a ^ er ^^ _ &e ious 

1959 ; Cruden 1965). It is unfortunate " ^^J^^fa 
excavationsintrhscmciallyimportantpartofthest^ P ^ 

final form, so that the cemetery ^ see Morris 

mediate vicinity of the chapel car i bariy ^ as ^ A 

to this period. 


parallels would suggest that it c ™' fthelater christian cemetery, 

aPictishsymbol-stone.Bemg placed The simp iicity of 
it would then presumably have had P^™f pearlier, pre-Norse, 
the designs of the cross-slabs ^^^^^that, apart 
rather than the later, ^an No^ 

the attribution of ^^^^ the earlier stone foundation is 
Viking period (see Mom 1982b). bqua y, 

precisely "^^^^^^S^mS a Christian Norse context 
is , despite equivocal datmg ^ ^d has recently accepted a 



well as demonstrably Pictisn or pre v s db JohnHunter andmyself 
further away from this later ecclesiastical focus. 

6 ° Bsv&ta 

the site as a whole had "special status", but, as Dr John Hunter has said, whether 
ifderivedfrompolitical considerations or monastic impetus is., .finely balanced 
according to current evidence" (Hunter 1986, 171). In this writer' s opinion, the 
balance is now firmly tipping towards the political. 

I have elsewhere discussed in detail the evidence from other sites in Orkney 
and Shetland brought forward by Dr Radford (Morris 1 990), and have argued 
that, despite the manifestly important stone shrines and Pictish silver treasure 
from StNinian's Isle, and similar shrines at Papil (Thomas 1973a; 1973b; 1983; 
MacRoberts 1965; Wilson 1969; 1973), the evidence for a monastic 
establishment at either site is questionable. Similarly, from my own excavations 
at the Brough of Deerness, I have concluded that there is an absence of firm 
evidence for such an establishment there at the pre-Norse period (Morris with 
Emery 1986). Indeed, I would argue that the evidence here, by analogy with 
other sites in the North Atlantic region, best fits with an explanation in terms 
of the Christian Norse period. By extension I have, therefore, raised the 
question as to whether the stone building that Thomas has hypothesised 
(Thomas 1973b, 11-14; 1983, 287-8) belowthe later stone chapel (itself oftwo 
phases) from StNinian's Isle could come from a later (i.e. Norse) period? In 
the light of evidence discussed from the Brough of Deerness, it might be asked 
whether this is not a more credible explanation of the evidence? It is also far 
from clear that the treasure from the site was necessarily from inside a building. 
(This does not preclude the possibility of a yet earlier building, perhaps of 
timber, unrecognised or unrecorded in the excavations of 1958). It would seem 
to me inescapable that the logic of the argument requires the earlier foundations 
at the Brough of Birsay to come from the same period; and to be seen perhaps 
in the context of a private chapel of the Norse Earl. I would argue now that the 
Early Christian date and 'Celtic' monastic model has become something of a 
straight] acket. 

Having asserted that the direct evidence for the Brough of Birsay as a 
monastic site of the so-called 'Celtic church' in the pre- Viking period is both 
questionable and, on the basis of parallels elsewhere, perhaps better explained 
in a later Norse context, I will now turn to the evidence for Western Britain, 
from where Dr Radford derived his 'model' . I think that we can now assert that 
the evidence, perhaps from Gateholm in Dyfed, and certainly from Tintagel in 
Cornwall, is better explained in terms of a secular power-centre rather than an 
ecclesiastical site. This does not lessen the importance of Birsay, but simply 

I 1 Britain^.heevid.nceforT^ 
shows it to be best explained. 
Tintagel and Western Britain, 

key sites to the exposition of the theme of The CI » Mon ry 
(Eiofotd 1962a). Despite cntiasms to toe 1970s (Bunow . 

1939; 1942; 1956, 1968, 19/3, v*»h „ and p ro f esS or 

church archaeologists such as I ** fes8 ^^ to 
Thomas (Thomas 1971a andb, but see Thoma ^ 0 Sme 1980s byDark 

( 1985),Thomas andFowler l^Thornasa, tQ ^ 

b; Thomas and Thorpe 19 88 Thomas ^ ; simultaneously , renewed 

"Deconstruction of a Monastery (Thomas p 

flfc 5.10. 2M«* : fC™«^W^ 0i« 

6 J Drvtcnn 

excavations both on the Island and the Churchyard sites in the 1990s have 
examined several aspects of the new approach to Tintagel and this recent work 
is beginning to be brought into the public domain via interhn reports (Morris, 
Nowakowski and Thomas 1 990 ;Nowakowski and Thomas 1991;Nowakowski 
and Thomas 1992; Morris with Emery and others 1991; Morris with Harry, 
Johnson and others 1993; Morris (ed) 1994; Harry and Morris (eds) 1995) and 
the first of the final reports on work on the Island is now available (Batey, 
Sharpe and Thorpe 1993). This is not the place in which to enter into a 
discusssion of the details of these excavations, which are ongoing, but it can 
certainly be stated that, for the present, Tintagel no longer provides a monastic 
'model' againstwhich other sites suchas the Brough of Birsaymightbejudged. 
We are beginning in this work to 'flesh out' a new interpretation of the island 
site as a secular stronghold of the early rulers of Cornwall (Fig. 3.10) but 
associated with an ecclesiastical focus in the area of the modern Parish Church. 
We also see it as intimately involved in trade and exchange with the 
Mediterranean world - as evidenced by imported pottery - and as remaining as 
a key site of the period, if in a radically different role. 

Brief reference has been made above to the reinterpretation by the author of 
the Brough of Deerness 
(Morris 1977; Morris with 
Emery 1986; Morris 1990) and 
we might also add that the 
fourth of the major sites 
brought forward by Radford 
for his Celtic monastic 'model', 
Gateholm in Dyfed, no longer 
appears to fit the requirements . 
Excavated hi 1909 (Cantrill 
1910) and 1929 (Lethbridge 
and David 1930), it shared 
many features with Tintagel as 
perceived in the inter- War and 
post-War years: isolated 
location, enclosing bank, 
groupings of rectangular huts. 
(Fig. 3.11) Work in the 1970s 
(Davies etal 1971) andarecent 

§? Britain 

comparisons between to l^ZXLZ^ The notion 
attribution as monastic s,te S , both m to pa*»m » „ 

i , f iin if nnt 130- Davies aZ 1971, Cantnn i^iu;, « 
least 110, it not uu. „ lirve v of Tintagel Island 



appear to share several common features. 

These Clarities would incline me ™ 
northern and western Britam m thrs penc-d w ^ h ^^ ouldusethe 

differences. M^^^ 10 ^^^^^^^ 
term "Pictish" to describe all those peep b ^f^^Z CO me down 

deriving from a differing language base, the adoption of the ogam tf* V 

% Britabn 

bone pins, antler combs etc) with a wider northern and western British "cultural 
orbit", and wemightwish to pointto a common concern with fine metalworking 
at places such as Brough of Birsay (Curie 1 982), Dunadd in Argyll (Campbell 
and Lane 1 993) and Dinas Powys (Alcock 1 963) in South Wales, all nominally 
high status sites of differing groups of "Celtic" peoples. 

In the past, the northern Pictish area has perhaps been neglected, almost 
marginalised, as attention has concentrated on Strathearn, Fife, Angus and the 
eastern Grampians and seaboard ("The Province of Mar": cf Simpson 1944), 
but the archaeological evidence from the Moray Firth area, Birsay in Orkney 
and St Ninian's Isle in Shetland clearly give strong evidence for a different 
perception. Within this northern area, Birsay Bay and the ' island' of the Brough 
within it would seem to have been an important place. Although there are 
differences in particular details between Birsay and Tintagel (not least in the 
range of external connections), they didperhaps have a similar role within their 
contemporary societies within Early Medieval Britain. Both were peninsulas 
or islands which appear to have functioned as power-centres for their immediate 
localities and, we presume, for the wider regions. The Brough of Birsay at one 
extremity of Britain now has a comparable contribution to make to our 
understanding of this period of protohistory, to that from Tintagel at the other 
end. It has been my privilege to work on both sites, and if I am now propounding 
different interpretations to those brought forward by my predecessor, Dr 
Radford, I am nevertheless only too well aware that all reinterpretation 
necessarily builds upon the work of those who have gone before and that Dr 
Radford already saw 'Scotland' in the wider framework of the British Isles in 
the Early Medieval period. 

Dept. of Archaeology 
University of Glasgow 

Acknowledgements-—- ~~ ~ ~~ ~~ — 

This paper represents an expanded version of the lecture given in the University 
of St Andrews in February 1 995 . 1 am grateful to Dr Crawford for the invitation 
to contribute to the Day Conference and to both her and the audience for their 
tolerance in relation to my illness during the lecture itself; I hope that the printed 
version is some recompense for the below par performance on the day! The 
lecture returned in part to a theme which was considered in my O'Donnell 

6 J Britain 

lecture in the University of Wales in 1993. Tins » currently ^ 
publ 1 cat 1 on 1 n»C^ C a > andtnere are inevitably axeas of overlap between 
C but the very different context of the two may perhaps provide some 
for ^repetition. I am grateful to Dr Colleen Batey for her help 
with discussion of the content of this text. 

References ' ~~~~ «•*"■■ 

Settlement in Glamorgan, Cardiff. 

Alcock, Leslie. 1984. A Survey of Pictish Settlement Archaeology, in FrieU & 
Watson (eds) 1984,7-41. 

Anderson, Alan O. (trans). 1922. Early Sources of Scottish History AD 500- 
1286, 2 vols, Edinburgh. 

Anderson, Alan O. and Anderson, Marjorie O. (ed & trans). 1961 , Adomnan s 
Life ofColumba, London 

Anderson, Joseph. 1881. Scotland in Early Christian Times, The Rhind 
Lectures in Archaeology, Edinburgh. 

Armit, Ian. 1990. Brochs andbeyondin the Western Isles, in Armit (ed) 1990, 

Armit, lan(ed).1990.BeyondtheBrochs.ChangingPerspectives in the Later 
Iron Age in Atlantic Scotland, Edinburgh. 

Ashe, Geoffrey (ed). 1968. The Quest for Arthur's Britain, London. 

Ashmore, Patrick J- 1980. Low cairns, long cists and symbol stones, Proc Soc 
Antiq Scot 110, 1978-80, 346-355. 

Batey, Colleen E. 1989. The Artefactual Assemblage in Morris 1989, Chap 7. 

Batey, Colleen E. and Moms, Christopher D. 1983. Part II: The Finds, in 
Hedges 1983, 85-108. 

66 Britain 

Batey, Colleen E., Sharpe, Adam and Thorpe, Carl. 1993. Tintagel Castle: 
archaeological investigations of the Steps are 1 989 and 1 990, Cornish Archaeol 
32, 47-66. 

Batey, Colleen E., Jesch, Judith and Morris, Christopher D. (eds). 1993. The 
Viking Age in Caithness, Orkney andtheNorth Atlantic. Select Papers from the 
11th Viking Congress, Kirkwall & Thurso 1989, Edinburgh. 

Bigelow, GeraldF. 1984. Two Kerbed Cairns from Sandwick, Unst, Shetland, 
in FrieU and Watson (eds) 1984, 115-129. 

Bigelow, GeraldF. 1985. Sandwick, Unst and Late Norse Shetland Economy, 
in Smith (ed) 1985, 95-127 

Bourke, Cormac. 1980. Early Irish Hand-bells, J Roy Soc Antiq Ireland 110, 

Bourke, Cormac. 1983. The hand-bells of the early Scottish church, Proc Soc 
Antiq Scot 1 13, 464-468. 

Breeze, David J. 1994. The Imperial Legacy - Rome and her Neighbours, in 
Crawford (ed) 1994, 13-19. 

Burrow, Ian C. G. 1 973 . Timagel - Some Problems, Scot Archaeol Forum 5, 99- 

Campbell, Ewan and Lane, Alan. 1993. Celtic and Germanic Interaction in 
Dalriada: the 7th-Century Metalworking Site at Dunadd, in Spearman and 
Higgitt (eds) 1993, 52-63. 

Cant, Ronald G. 1983. Introduction, Orkney Heritage 2, 7-12. 

Cantrill, T. C. 1910. The hut-circles on Gateholm, Pembrokeshire, Archaeol 
Cambrensis LXV, 271-82. 

Carter, StephenP., Haigh, David, Neil, Nigel R. J., and Smith, Beverley. 1984. 

6J Britain 

Interimreport onthe structures atHowe, Stromness, Orkney, Glasgow Archaeol 
J 11, 61-73. 

Close-Brooks, Joanna. 1980. Excavations in the Dairy Park, Dunrobin, 
Sutherland, 1977, Proc Soc Antiq Scot 110, 1978-80, 328-245. 

Close-Brooks, Joanna. 1984. Pictish and other Burials, in Friell and Watson 
(eds) 1984, 87-114. 

Close-Brooks, Joanna. 1989. Pictish Stones in Dunrobin Castle Museum, 

Colgrave, Bertram & Mynors, Richard A. B. (ed & trans). 1981. Bedes 
Ecclesiastical History of the English People, Oxford. 

Cramp, Rosemary J. 1976. Monastic Sites, in Wilson (ed) 1976, 201-52. 

Crawford, BarbaraE. 1983. Birsay and the Early Earls andBishops of Orkney, 
Orkney Heritage 2, 97-118. 

Crawford, Barbara E. 1987. Scandinavian Scotland, Scotland in the Early 
Middle Ages 2, Leicester. 

Crawford,BarbaraE(ed). \994.ScotlandinDarkAgeEurope, StJohn'sHouse 
Papers No 5, St Andrews. 

Crawford, Barbara E (ed). 1995. Northern Isles Connections. Essays from 
Orkney and Shetland presented to Per Sveaas Andersen, Kirkwall. 

Crawford, Iain A. 1974. Scot (?),Norseman and Gael, Scottish Archaeological 
Forum 6, 1-16. 

Cronyn, Janey M., Morris, Christopher D., and Owen, Olwyn A. 1984. A 
Decorative Mount from the Brough of Birsay, Orkney: A preliminary Note, 
Univs of Durham & Newcastle Upon Tyne Archaeol Reps for 1983 7, 55-60. 

Cruden, Stewart H. 1958.Earl Thorfmn the Mighty andthe Brough ofBirsay, 
inEldjarn(ed) 1958, 156-162. 

<£ Britain 

Cruden, Stewart H. 1 965. Excavations at Birsay, Orkney, in Small (ed) 1 965 

Curie, Cecil L. 1940. The Chronology of the Early Christian monuments of 
Scotland, Proc Soc Antiq Scot 74, 1939-40, 60-1 16. 

Curie, CecilL. 1974. An engraved lead disc from the Brough ofBirsay Orkney 
Proc Soc Antiq Scot 1 05, 1972-4, 301-7. 

Curie, Cecil L. 1 982. The Pictish and Norse Finds from the Brough ofBirsay 
1934-74. Society of Antiquaries of Scotland Monograph Series 1, Edinburgh 

Curie, Cecil L. 1 983 . The Finds from the Brough ofBirsay 1 934- 1 974 Orkney 
Heritage 2, 67 '-81 ' 

Dark, Kenneth R. 1985. The plan and interpretation of Tintagel, Cambridge 
Medieval Celtic Stud 9, 1-18. 

Davies, Jeffrey L., Hague, Douglas B. and Hogg, A. H. A. 1971. The hut 
settlement on Gateholm, Pembrokeshire, Archaeol Cambrensis 120, 102-10. 

Dumville, DavidN. 1 976. A Note on the Picts in Orkney, Scot Gaelic StudXE, 

Edwards, Nancy and Lane, Alan. 1988. Early Medieval Settlements in Wales 
AD 400-1000, Bangor/Cardiff. 

Eldjarn, Kristjan (ed). Third Viking Congress, Reykjavik 1956, Reykjavik. 

Fairer, James. 1862. The Knowe of Saverough, Gentleman 's Magazine 213, 

Fenton, Alexander and Palsson, Hermann (eds). The Northern and Western 
Isles in the Viking World. Survival, Continuity and Change, Edinburgh, 

Forsyth, Katherine. forthcoming a. The Ogham-Inscribed Spindle Whorl from 
Buckquoy: Evidence for the Irish Language in pre-Viking Orkney?, Proc Soc 
Antiq Scot. 

S Britain 

Th^Oehain Inscriptions of Scotland: An 

Forsy fr, ferine. *£^?^W*I*B^^ 
Bdited Cor^s, nnpub Ph. D. thesis, v V 

« i c-^o ^ 1 984 ficft'sfe Studies. 
MA 3. Gen , P , BAR Brit Ser 125, 

Sett/emenf, Burial amUrf m L>a/*^ 

their immediate predecessor, m Fenton an 

^s- r —- ~°— 

Friell and Watson (eds) 1984, 131-4 

Lewis, in ArmrtCed) 1990,71-107. 

. «, ,9X3 TrialexcavationsonPictiahandVadngsettlem^a 

& Microfiche 40-102 

„ w 1Q87 BM GMrn ^ a ^* e 5ro C fo 0 /0^y,3vols,BARBnt 

Vi 1 1958 The Onghr Centre of the Pictish Symbols Stones, 
Henderson, Isabel. 195* TteU g 
Proc Soc Antiq Scot XCI, 195 / a, t, Isabel. 19, 1 . No,* Picriand, ,n Meldtan, (ed) 1971,37-52. 


Henderson, Isabel. 1975. Inverness, A Pictish Capital, in Inverness Field Club 

Henderson, Isabel. 1990. The Art and Function of Rosemarltie's Pictish 
Monuments, Groam House Lecture, Rosemarkie. 

Hunter, JohnR. 1982. Glass: A Report, in Curie 1982, 46-7. 

Hunter, John R. 1983. Recent Excavations on the Brough of Birsay, Orkney 
Heritage 2, 152-170. 

Hunter, JohnR. 1986. Rescue Excavations on the Brough of Birsay 1974-82. 
Society of Antiquaries of Scotland Monograph Series 4, Edinburgh. 

Hunter, John R. 1990. Pool, Sanday: A Case Study for the Late Iron Age and 
Viking Periods, in Armit (ed) 1990, 175-197. 

Hunter, John R. and Morris, Christopher D. 1982. Appendix: Excavation of 
Room 5, Brough of Birsay, Clifftop Settlement, 1973-4, in Curie 1982, 124- 

Jackson, Kenneth H. 1955. The Pictish Language, in Wainwright (ed) 1955, 

Inverness Field Club. 1975. The Hub of the Highlands. The Book of Inverness 
and District, Edinburgh. 

Jackson, Kenneth H. 1977. Appendix 6. The ogam inscription on the spindle 
whorl, from Buckquoy, Orkney, in Ritchie 1977, 221-2. 

Kaland, Sigrid H. H. 1993. The Settlement of Westness, Rousay, in Batey, 
Jesch and Morris (eds) 1993, 308-317. 

Lamb, Raymond G. 1 993 . Carolingian Orkney and its transformation, inBatey, 
Jesch and Morris (eds) 1993, 260-271. 

Lamb, Raymond G. 1995.Papil, Picts and Papar, in Crawford (ed) 1995,9-27. 

Ti Britain 

Lane, Alan. 1988. Gateholm, in Edwards and Lane (eds) 1988, 72-5. 

„ a H E 1930. Excavation of a house-site on 

Nortfe™ Swi », 25-30 [repnnted m ,« Celuc 
Laing, BAR Brit Ser, Oxford, 25-30]. 

Arltmr 1974 The B.och of Banian, North Ronaldsay, Orkney, 

M_, D a„d. 1965 — -a, — " 

Isle treasure, in Small (ed) 1965, 224-246, 

»a m ,;o to C.1974.Tne N or« k er„ F r„r > rier after AD369, Gto ^*cW 
J 3, 34-42. 

Marwick, Hugh. 1952. Orkney Farm Names, Kirkwall. 


Small (ed) 1987, 31-44. 

^rctoeofogia Atlemtica 2, 65-79. 

o+ -Riroav 1 979 Northern Stud 16, 

Moms, ChristopherD. 1980.E— 
17-28, and t/mvs o/Dur/wm and Newcastle Upon ly 

u n, QR1 Excavations at Birsay, Orkney, 1980. Untvs of 
• mm The Vikings in the British Isles: some aspects 

Chichester, 70-94. 

« Brttatrv 

Morris, Christopher D. 1982b. Excavations at Birsay, Untvs of Durham & 
Newcastle Upon Tyne Archaeol Reps for 1981 5, 46-53. 

Morris, ChristopherD. 1983. Excavations around Birsay Bay, Orkney, Orkney 
Heritage 2, 119-151. 

Morris, ChristopherD. 1985. Viking Orkney, in Renfrew (ed) 1985, 210-242. 

Morris, ChristopherD. 1989. The Birsay Bay Project Volume 1 . Brough Road 
Excavations 1976-1982, University of Durham, Department of Archaeology, 
Monograph Series 1, Durham. 

Morris, Christopher D. 1990. Church and Monastery in the Far North. An 
Archaeological Evaluation. Jarrow Lecture 1989. Jarrow. 

Morris, ChristopherD. Forthcoming. Brough of Birsay, Orkney. Excavations 

Morris, ChristopherD. (ed). 1994. Tintagel Castle Excavations /PP5by Rachel 
C. Harry and Paul G. Johnson, English Heritage/Univ of Glasgow. 

Morris, Christopher D., Nowakowski, Jacqueline and Thomas, A. Charles 
1990. Tintagel, Cornwall: the 1990 excavations, Antiquity 64, No 245, 843-9. 

Morris, Christopher D. with Emery, Norman. 1986. The chapel and enclosure 
on the Brough of Deerness, Orkney, survey and excavations, 1 975-7, Proc Soc 
Antiq Scot 116, 301-374 & Aches 2-4. 

Morris, ChristopherD. with Emery, Norman. 1991. Tintagel Castle Excavations 
1990, English Heritage/Univ of Durham 

Morris, Christopher D. with Harry, Rachel C. and Johnson, Paul G. 1993. 
Tintagel Castle Excavations 1991, English Heritage/Univ of Glasgow. 

Morris, John (ed & trans). 1980. Nennius' British History and the Welsh 
Annals, Chichester. 

Neil, Nigel R. W. 1985. Excavations at Howe, Stromness, in Renfrew (ed) 
1985, 205-7. 

t5 Britatn 

Interim report, Truro. 

, „ a rwles 1992. Grave News from 

Churchyard, Cornwall 1991, Truro. 

, ^ i twW V fedsrt 1983. From the Stone Age to the 
O'Connor Anne and Clarke, David V. (easy. i*o 

^ 2UJi Stevens Edinburgh. 

Palsson, Hermann and Edwards, Paul (trans). 1978. Orkneyinga Sa g a. The 
History of the Earls of Orkney, London. 

Brit Ser 102, Oxford 

Pebie, George. 19,3. Nobee of .he Broehs or Round Towers of Orkney, 
ircfcaeo/ScofV, 71-94. 

Radford, C. A. Ralegh- 1935. Tnrtageh the Casbe and Celhc Monaster,. 
Interim report, Antiq JXV, 401-19. 

padford, C. A. Ralegh. 1942. Tin.agel HW-U^ ' ^ *- 
Lnwafl LXXXVI (NO 86) - Appendnc for 1942, 25 41. 
Rad f ord,C.A.Ra.egh.l956. Im po rt edP„«e,yfo m da,T 1 Pfage,,Corawal.,i. 

Harden (ed) 1956, 59-70. 

Radford, CA.Ra.egh. .959. B. ** «— 
of Sirsoy. Official Guide, Edinburgh. 

Katford, C. A. Rule*. «62a. The C.lbo Monaster in BMi^ 
Cambrensis CXI, 1-24. 

Radford, C. A. Ralegh. 1962b. Art and Arehheeture: Gelfie and Nor,., in 
Wainwright(ed) 1962, 163-187. 

7 * BtfliSto 

Radford, C. A. Ralegh. 1968. Romance and Reality in Cornwall, in Ashe (ed) 
1968, 59-77. 

Radford, C. A. Ralegh. 1973. Summary and Discussion, Scot Archaeol Forum 
5, 136-40. 

Radford, C. A. Ralegh. 1 975 . Tintagel, in Radford and Swanton (eds) 1 976, 1 6- 

Radford, C. A. Ralegh. 1 983. Birsay and the Spread of Christianity to the North, 
Orkney Heritage 2, 13-35. 

Radford, C. A. Ralegh and Swanton, Michael J. (eds). 1975. Arthurian Sites in 
the West, Exeter. 

Renfrew, A. Colin (ed). 1985. The Prehistory of Orkney BC 4000-1000 AD. 

Ritchie, Anna. 1972. Painted pebbles in early Scotland, Proc Soc Antiq Scot 
104 (1971-2), 297-301. 

Ritchie, Anna. 1974. Pict and Norseman in Northern Scotland, Scot Archaeol 
Forum 6, 23-36 

Ritchie, Anna. 1977. Excavation of Pictishand Viking-age farmsteads at 
Buckquoy, Orkney, Proc Soc Antiq Scot 108, 1976-7, 174-227. 

Ritchie, Anna. 1983. Birsay Around AD 800, Orlmey Heritage 2, 46-66. 

Ritchie, Anna. 1985. Orkney in the Pictish Kingdom, in Renfrew (ed) 1985, 

Ritchie, Anna. 1986. Brough of Birsay. Official Guide, Edinburgh. 

Ritchie, Anna. 1989. Picts. An Introduction to the Life of the Picts and the 
Carved Stones in the Care of the Secretary of State for Scotand, Edinburgh. 


Ritchie J N Graham. 1969. Two New Pictish Symbol Stones from Orkney, 
Proc SocAntiq Scot 101, 196Z-9, 130-3. 

J994, RCAHMS, Edinburgh. 

o/Orte^ ani 3 vols, Edinburgh, nos 1 & 6, 1-5 & 7. 

Simpson W. Douglas. 1944. The Province of Mar, Rhind Lectures in 
STtogy 1941, Aberdeen Univ Stud 121, Aberdeen. 

Skene, William F. 1867. Chronicles of the Picts and Scots, Edinburgh. 

Skene, William F. 1876. Celtic Scotland, 2 vols, Edinburgh. 

its Treasure, Aberdeen Univ Stud 152, Oxford. 

Small Alan (ed). The Fourth Vikin g Congress, York, August 1961, Aberdeen 
University Studies 149, Edinburgh & London. 

Small, Alan (ed). 1987. The Picts. A Ne W Look at Old Problems, Dundee. 

Smith Beverley. 1990. New Insights into Later Iron Age Settlement in the 
North: Howe, m Armit (ed) 1990, 32-40. 

^tthBeverleyBallin(ed).1994.H OWe .Fo M rMZ/ e nia 0 /Orte e yPr e W 5 to^, 

Smith, Brian (ed). 1985. Shetland Archaeology. Ne» Work in Shetland in the 
1970s, Lerwick. 

Smyth, Alfred?. 1984. Warlords andHolymen. Scotland AD 80-1000,^ 
History of Scotland 1, London. 

Spearman, R. Mrchael and Higgitt, John. (eds). 1993. The Age of Migrating 

7 S Britam 

Ideas. Early Medieval Art in Northern Britain and Ireland, Edinburgh and 

Sterckx, Claude. 1970. Les jeux de damiers celtiques, .4w«a/es de Bretagne 77 
597-609. S 

Sterckx, Claude. 1 973a. Les trois Damiers de Buckquoy (Orcades), Annales de 
Bretagne 80, 675-689. 

Sterckx, Claude. 1973b. Les jeux de damiers celtiques, Etudes Celtiques 13 

Taylor, Albert B. (ed). 1938. The Orkneyinga Saga: a new translation with 
introduction and notes, Edinburgh & London. 

Thomas, A. Charles. 1971a. The Early Christian Archaeology of North Britain 

Thomas, A. Charles. 1971b. Rostat, Rosnat and the early Irish Church Eriu 

xxn, 100-6. 

Thomas, A. Charles. 1 973a. Bede, Archaeology and the Cult of Relics. Jarrow 
Lecture 1989, Jarrow. 

Thomas, A. Charles. 1973b. Sculptured stones and crosses from St Ninian's 
Isle and Papil, in Small, Thomas and Wilson 1973, 8-44. 

Thomas, A. Charles. 1982. East and West: Tintagel, Mediterranean imports 
and the Early Insular Church, in Pearce (ed), 17-34. 

Thomas, A. Charles. 1983. The Double Shrine'A' from St Ninian's Isle, 
Shetland, in O'Connor and Clarke (eds) 1983, 285-92. 

Thomas, A. Charles. 1988a. Tintagel Castle, Antiquity 62, 421-434. 

Thomas, A. Charles. 1988b. The Context of Tintagel. A New Model for the 
Diffusion of Post-Roman Mediterranean Imports, Cornish Archaeol 27, 7-24. 


Da %qe 
V Britain, 

Thomas A. Charles. 1993. Arthur and Archaeology, London. 

Thomas, A. (ed, I» Tintagel Paper, - (feri* « « 

Annual Review 1984-5, 16-21. 

^nvr Catalogue of all non-Medieval 

tttt^^ffi* r Stod -* EGgllsh 

Sage, unpub report on file in Truro and London. 
Thomson, WilhamP. L. 1987. History of Orkney, Edinburgh. 
Wainwright^rederickT. 1962. Picts and Scots, in Wamwright (ed) 1962,91- 

Wanawngh,, Frederick, (ed). 1955. * ProUe, of, he M Edlnbnrgn * 

W^ght,F ^ ed m *T.(ed).1962.^^-«-•E*^ 1 • &LO ■' d<, - 1 ' 
c ,osn FxcavaoonofasettlemmtandsoutemfaatNewmill, 

Wadnns, Trevor F. Wh f w- ™ *» — " ^ 

archaology, in Friell & Watson (eds) 1984, 63-86. 

Wdson, DavidM. ,969. M«- - »— * 


Aberdeen Univ Stud 152. Oxford 

Wilson, Davd M. (ed). 1976. The ArcHaeo.ogy 

*r i. IMA trans) 1978. Gildas. The Ruin of Britain, 
Winterbottom, Michael (ed & trans;, if/ 


T§ BtAtaun. 


Trade m tke Dark- Age West a peripheral activity? 

Ewan, Campbell 

1. Introduction, 

At the 1993 Day Conference 'ScotlandinDark-Age Europe' (Crawford, 1994), 
Alan Lane presented some of the basic results of my research into the trade and 
imports in the Celtic West (Lane 1 994), so this paper will be more speculative 
and wide-ranging in an attempt to assess the importance of trade in the period 
AD 500-800, and its relative importance to Scotland in relation to the rest of 
western Britain and Ireland. 

The 'peripheral' nature of trade which I wish to re-examine has two aspects: 
the geographical sense, especially in terms of core/periphery models; and the 
socio-economic sense, relating to the relative importance of trade in the 
maintenance of social relations. 

My reasons for re-assessing the importance of trade derive from a number 
of recent articles and books which argue that some archaeologists have 
overemphasised the importance of long-distance trade, and that most of the 
demonstrable exchange was of fairly minor economic significance and not 
instrumental in social change (Mytum 1992; Griffiths 1992; Wooding 
forthcoming). Wooding in particular puts forward a model of trade by coastal 
tramping, with ships following random routes, driven by the needs of cargoes 
and clients they encounter in various ports, (which I would paraphrase as the 
Para Handy model). Another aspect of recent criticism is the suggestion that 
the scale of the finds is not consistent with regular trade, the numbers of 
imported pottery vessels being compared unfavourably to those found in some 
mercantile contexts in the Mediterranean, for example the seventh-century 
Yassi Ada wreck (Alcock and Alcock 1990, Alcock 1995). 

No one doubts the importance of the imports in providing a chronological 
framework for the region, there being no other independent dating mechanism 
for most sites in this area; or the fact that they illustrate some sort of long- 
distance contact and brought some exotic objects to Insular sites, but it is the 
scale and significance of the activity which is disputed. 




lit me importance of the trade; and end with some speculation about the 
processes involved. 

2. Summary of Import groups 

■lo^t 1 of Wthom (Campbell formcoming b) and d.soovcnes m 

Th, Utar Continental system can now be seen to also have two phases The 

hlnbeing trade items mtheir ownright, as Thonrashas 
TV ffi^nalvsis has shown traces of non-native red dyestuffs (Dyer s 
Scientific analysis has f °^ ted onthe basis of contemporary 

spices may nav ^ recent excavatlons at 

spice coriander and , hehe dill hav ^ ^ ^ 

SKS. SI of any cargo was probably salt and 
TMedherranean imports, around the Irish Sea and western Scotland. 

8 ° Britain, 

3. Trade or not? 

The criticisms outlined in the introduction lead me to reassert my belief that the 
distribution patterns shown by the imports are the result of sustained, regular, 
directed trade, (in other words mercantile trade), rather than one of the 
alternative options, such as the tramping model of Wooding. Other minimalist 
explanations include that of Charles Thomas (1990), who has suggested that 
most of the Mediterranean imports could be the result of one, or very few, 
voyages of merchant entrepreneurs. This can be described as the Columbus 
model , where a merchant adventurer sets out into the unknown in the hope of 
discovering unknown riches. Another view is that the imports could be the 
chance keepsakes of returning travellers, such aspilgrims, tourists ormercenaries. 
Professor Vera Evison (1979) has suggested this explanation for the sixth and 
seventh century Anglo-Saxon pottery imports to Kent, in what might be called 
the duty-free bottle model. 

As far as the Mediterranean imports are concerned the date of the wares can 
now be ascertained quite closely from work at Carthage and elsewhere (Fulford 
& Peacock 1984). The minimum period covered by the different tableware 
forms found in Britain is a generation, with some vessels that can be dated to 
around AD 500 and others to AD 525/35, but the likely span is more, probably 
50 years from 500-550. This spread of dates in itself shows continuing contact. 
More convincing evidence comes from the study of the Mediterranean patterns 
of trade, which shows that Phocaean Red Slipware was restricted to the eastern 
Mediterranean except in the period AD 475-550, when its range expands 
dramatically with exports to Italy, Spain and Portugal (Hayes 1974, maps 3 1- 
4). The British distribution is quite clearly an extension of this trade and of the 
same character, with the same package of wares, and should date over the same 
range of time. Furthermore, it is difficult to see why a round voyage of 1 0,000 
km from Byzantium should be undertaken on the off-chance that some 
significant trade goods would be available on arrival in Britain. A cargo load 
of tin, taking the Yassi Ada ship as a model (Bass & van Doornick 1 984), would 
be around 40 tons. This would represent a large commitment of effort and time 
to mine and refine on the part of the British, an effort unlikely to be undertaken 
without a guarantee that the Byzantine merchants would appear, even if only 
once a year. Of course the same applies in reverse to the merchants. Although 
it was short-lived, I would suggest this was a true trading venture, though 
whether it was undertaken by entrepreneurial merchants in a period of lax 

I 1 Britain 

imperial control (Fulford 1989), or by imperial procurators of supplies, could 
be debated. 

and the same applies to the distinctive decorated glass ^chdjers taati* 
of Northern France (Fig. 4.3) (Campbell forthcommg o) The were ^many 

wecTrlg^ Insular areas. Forcompariso^atMiddleSaxonil^csorne 

tovLus parts of north-western France and Belgium (Trmby 19SS). New 
evidence from Whithorn, the first western site to have a go od start tod 
Iquence of deposits, shows the imported pottery continumg throu J most o 
8 "b Wses deposited between AD 500 and 700, which must indicate tha 
ei shipments of Mediterranean and Continental wares were amvrng 
rcCb^orthcomingb). Ag^a it seems clear ihat ^n^^ 

years, suggesting 
mercantile trade aimed 
at satisfying a particular 
need for commodities 
available in these areas. 
The nature of these 
commodities is un- 
known, thoughthey must 
have been organic, as 
surviving artefacts of 
Insular origin are almost 
unknown inFrance atthis 
period. Suggestions have 
included slaves, leather 
goods, furs, and perhaps 
even cereals. 

One argument against 
the distributionbeing the 


D %v qe 

8 J Britain 

Fig 4 1 Distribution ofPhocaean Red Slipware in western 
Europe. Size of symbol is proportional to number of vessels. 

result of trade is the scale of the finds. The total number of vessels so far known 
is relatively small though it is growing rapidly from the current tally of 300 
Mediterranean and 500 Continental. However, it is not suggested that in either 
system the main item of trade was the pottery, which may have been incidental 
to the main cargo of perishable goods. In many known Mediterranean wrecks 
pottery is only a minor element of mixed cargoes (Parker 1984). Even so, the 
British numbers compare well with those from western Mediterranean sites 
(Fig. 4. 1). It is also perhaps instructive to look at another example of imported 
pottery on western sites, namely the thirteenth- andfourteenth-century Saintonge 
ware from the same region of south-west France as the sixth/seventh century 
imports, and itself certainly the product of mercantile trade. The total numbers 
of vessels and sites until recently was similar to that of E ware, but John Allan 
has estimated, using later Exeter PortBooks statistics, that millions of Saintonge 
pots may have been imported, and that urban excavations may be revealing 
only one thousandth of the imported amount (Allan 1983). While the figures 
cannot be taken at face value, this does illustrate that small numbers of pots may 
be the only visible residue of a much more significant trading system. 

4. Socio-economic context of the import sites 

How can we assess the socio-economic importance of this trade? The 
archaeologist has only two options, to analyse where and in what context the 
imports are found, and to look at patterns over time. 

35 " 

30 " 

25 " 





15 " 

10 - 

5 " 

0 - 

Continental pottery E ware 

V 2 1 3 1 4 1 5 ' 6 1 7 1 8 ' 9 ' 1 0 'TT' 12' 1 3 1 14' 15 ' 16' 17 1 1 8 1 1 9 l 20'21 , 22 l 23 l 24 l 25 1 26 l 
No of vessels 

Fig. 4.2. Relative frequency of imported E ware on Insular sites, shown by 
number of vessels per site. Dunadd is the site with most vessels. 

§2 Britain 

141). It-PO-^^^tweTItus sites in then hinterland. In 
imp ortation, with re ^ b ^ 

Ireland* is evenpossible inplace* > *^ ^ jor sites „ exatmned it can 
(Campbell 1991). If the characterises ° ^ ded 3 centres , UI1 defended centos 
be seen thatthere are three types of site. J^^ed ceG tres I want to look 

atindetail. Theseha^ 

other types of site (Tabled)- They ™ erseUkmoft i m ber-lacedrarnpart 
labour-.forexampleatCadburyCasa^ ^^oXr^ctedsticofwarrioranstocrats, 

arefound. Precious metals, which tnsM production of 

of the upper levels of society ar S rf ^ ^ mch 
personal jewellery. Lnparticu ortance as ameans of displaying 
^(1992)h-^^^SSoL«l«y evidence for royal 
and controlling rank. Finally, several have . g ^ 

tls. Given the gaps in the 

without documentary evidence for roy ft ,, nctiono f these central 

de^^^afPf^^^JStc group and reinf orcing 
access to exotic goods, ^^^oe resources. But are the imports 
S odalrelationsm P stfc^ 

central to this? Were they oneofmany suchso for tg 

0 the development of ---^ 
maintenance? r^c^^^^^ 

important, by looKing at w r ^ ^ &g mam 

As far as the Mediterranean .system is ^ntm Cornwall became a 
fortified centres had been abandoned, ^ se ^ leastbythe seventh century 

(Preston-lones andRose 1986). Tta «^ glo -Saxon advance, whh 
Uers r7 e. = 
the Battle of Dyrham m d / 


Table 4. 1 : Characteristics of major import sites 




gold/ silver 







Defended sites 









Mote of Mark 






































Dinas Powys 







Undefended sites 

South-west Britain 












Trading sites 

Cont.: Continental Med.: Mediterranean 

& £ Britain. 

this is stretching the evrdence for Cadbury ^Castte and ^omt^y (Rahtz 
as there is little evidence ot Saxon semeiucu 

lies in an area not 
& Watts 1989), and it is patently absurd for Tmtagd_ which ^ 

settled by Saxons until the tenth century. I * Economic 
Mediterranean market for metals caused ^^^Hke me cash crops 

this power vacuum. 

The p,«ure regarfcg the Coital import .system 

Longbury and Garranes. Some, such as w nun prov ince. 
into the Anglo-Saxon orbit and are ^^^^ J Dunadd 
But some such as ^^^£Sto the south- 
even expands in the eighth century. Tins sugg es ™ ^ ^ c b with 

western sites, some of these ^^^^^^^ external 
o +~ ntlipr forms of wealth which was not dependent soicjj u 

social control using other mechanisms or goods. 

There axe some clues to what might have been going on into north ^ 

types, ringforts and crannogs, the ^^^^ t0 bul ld. 

Whatever the cause, and 
Myrum (1992) has suggested 
the introduction of new 
technologies such as coulter 
ploughs and water-driven 
horizontal mills, there seems 
to have been increased 
production, with surplus 
directed into building and 
also fine metalwork. This 
surplus may have been the 

of the new power centres, 
Fig^.Reconstructionofglassves.elswUhopaquewhite digturbing the economic 
trailed decoration from Dinas Powys (left) and Whithorn Qf a abased 

Priory (right). 



system where small surpluses were absorbed and redistributed between lords 
and clients. The timely appearance of foreign merchants may have served as 
auseful outlet for the new wealth while increasing the power of the contemporary 
kings. The next few centuries saw a gradual agglomeration of power into larger 
political units in Ireland, though the lack of dynastic succession made it very 
difficult for power to be inherited in a stable way. The seventh century also saw 
the rise of the influence of the Church and an increasing involvement in secular 
politics. The Church was a prime consumer of surplus, both human and 
material, as the gifts given to it by the wealthy, unlike in secular society, did not 
have to be reciprocated, as the reward was postponed to the afterlife. In return 
rulers may have had their power legitimised by the Church. We can see this 
process at work in Adomnan's account of the relations between Aidan and 
Columba. When the Continental trading system decreases in the later seventh 
century, some of the surplus, which had been in effect siphoned off to the 
Continent, was probably invested in spiritual returns with the Church, and I 
would suggest that this accounts for the huge changes between 600 and 700 in 
church art, sculpture and metalwork, although undoubtedly some surplus went 
into personal secular ostentation. 

5. wider aspects 

If we look again at the hierarchy of import sites some interesting figures appear, 
with a gradual concentration of imports further north towards Scotland as we 
move from the sixth to seventh centuries. This is illustrated if we look at the 
five most important import sites, in terms of vessel numbers, in successive 
periods (Table 4.2). In purely geographical terms this is difficult to explain as 
one would expect the sites closer to the core economic region to have the larger 
import assemblages. It suggests that there came to be greater rewards in the 

Table 4.2: top five sites in terms of imported vessel numbers 








Cadbury Congresbury 


Cadbury Congresbury 

Dinas Powys 

Dinas Powys 

Dinas Powys 

Longbury Bank 


Cadbury Castle 

Mote of Mark 

Mote of Mark 

~ Britain 

^MortheF^sh—sbythe seven— 

factors presumably remained constant ^^^^Soofl^ 


If we compare tbe economic f^jfJ^^SS 
Celtic West and Anglo-Saxon England ^^J^Xelopment towards 
distance trade and craft specialisation, but lacks J ortowns 

^^^^^fff^^^^!^ seems to be 

directly InEnglandtradingplaces 
fromtheroyalsitesandundercontrololroyair a1though others have 

appear around 700 with the development o^^ 

across the Channel. 

TrademtheWest was indeed^ 

Mediterranean and Continental ^Z^ToooZl Ms hoped that this 
developments*— soclo -economic level, 

paper has shown that trade was slg T, Qver socia i development, 

however, and may have had that the northern 

As far as Scotland in Britain is conee rned, 700thano1iie rs 
togdornswereeconomicaUymoreadvan ^f^Z nAlhan itse] f 
furL south, something which the ^^^ZJrly development 

Department of Archaeology 
University of Glasgow 


8 i Brltto 

Ackno wlectcjemerits ' 

I would like to thank Anne Crone (AOC Edinburgh) for the information on 
plant remains from Buston crannog. 

References ~~ — . — ■ ~~, — 

Alcock, L & Alcock, E A 1990 "Reconnaissance excavations on Early 
Historic fortifications and other royal sites in Scotland, 1 974-84: 4, Excavations 
at Alt Clut, Clyde Rock, Strathclyde", 1974-5', Proc. Soc. Antiq. Scot. 120,95- 

Alcock, L 1995 Cadbury Castle, Somerset: The early medieval archaeology 

Allan, J. 1983 "Some post-medieval documentary evidence for the trade in 
ceramics", in Davey, P & Hodges, R (eds) Ceramics and trade, Sheffield, 37- 

Bass, G.F. &vanDoornick,F.H. 1982 YassiAda, a seventh-century Byzantine 
shipwreck (Texas) 

Broun, D 1 994 "The origin of Scottish identity in its European context", in 
Crawford 1994, 21-31 

Campbell, E 1991 Imported goods in the early medieval Celtic West: with 
special reference to Dinas Powys, (Unpublished PhD thesis, University of 
Wales, College of Cardiff). 

Campbell, E forthcorning a "The archaeological evidence for contacts: imports, 
trade and economy in Celtic Britain AD 400-800", in Dark, K R (ed), External 
contacts and the economy of Late Roman and Post-Roman Britain (W oodbridge) 

Campbell forthcoming b "Imported pottery and glass from Whithorn" in Hill, 
P Excavations at Whithorn Priory 

Campbell, E forthcoming c "A review of glass vessels in western Britain and 
Ireland AD 400-800" in Price J (ed) Glass in Britain, AD 350-800 (London) 

5? Britain 

Crawford, B E (ed) 1994 Scotland in Dark Age Europe (St Andrews; 

Evison,VI 1979 A corpus of wheel-thrown pottery in Anglo-Saxon graves 

Fulford M G. & Peacock, D.P.S. 1984 Excavations at Carthage: the British 
IS Vol. K2) TheAvenueduPresidentHabibBou^ba, Salammbo. the 
pottery and other ceramic objects from the site (Sheffield) 
Fulford M G 1989 Byzantium andBritainiaMeditmaneaap^p^ve on 

Archaeol. 33, 1-6 

Griffiths D 1992 "The coastal trading ports of the Irish Sea", in Graham- 
SS5u(ed) ViMng Treasures from the North-West, Liverpool, 63-72 

Hayes, J W 1972 Late Roman pottery (London) 

Holden, T 1995 "The waterlogged plant remains from Buston crannog", 
(unpublished AOC report, Edinburgh) 

Lane A 1994 "Trade, gifts, and cultural exchange in Dark-Age Western 
Scotland", in Crawford 1994, 103-115 

Mytum,H 1992 The origins of Early Christian Ireland (London) 

NiekeMR 1993 "Penannular and related brooches:secular ornament or 

Las. Early Medieval Art in Northern Britain and /re W, Edinburgh, 128 

Parker, A J 1984 "Shipwrecks and trade in the Mediterranean", Archaeol. 
Review Cambridge 3, 99-114 

Preston-Jones, A & Rose, P 1986 "Medieval Cornwall", Cornish Archaeol. 
25, 135-185 

Rahtz, P.& Watts, L. 1989 "Pagans Hill revisited", Archaeol J. 146,330-371 

s ° Britain 

Thomas, C. 1990 '"Gallici nautae de Galliarumprovinciis' - A sixth/seventh 
trade with Gaul, reconsidered", Medieval Archaeol. 34, 1-26 

Timby, J.R. 1988 "The Middle Saxon pottery", in Andrews, P. (ed), 
Southampton finds, volume 1; the coins and pottery from Hamwic, Southampton 
London,73-124 . 

Wooding, J forthcoming Communication and commerce along the western 
sealanes AD 400-800 Oxford 

U Britain 

?? Britain 

Place-rtames arui tke Early Ckurck m Eastern Scotlan,d 

oimon Taylor 

There is a rich seam of ecclesiastical place-names throughout Scotland, which 
manifests itself in elements such as annaid, cill, diseart, *egles, tearmann and 
kirk (1). We are fortunate in that many of them have been subjected to careful 
analyses by a host of scholars, such as Geoffrey Barrow-, Aidan MacDonald, 
John MacQueen, W.F.H. Nicolaisen, W.J. Watson and most recently Thomas 
Clancy (2). Nevertheless, we are still far from exhausting this seam, and in this 
paper I want to try to howk out a little more from it. I intend to concentrate on 
two elements in particular. One is the 'weel-kent' Gaelic element cill 'church'; 
the other is the element both. This is equally 'weel-kent', but in a pastoral- 
transhumant context rather than in an ecclesiastical one, since it is the Gaelic 
word for 'hut, sheiling', or 'bothy'. 

I have just completed a PhD on the settlement- and parish-names of Fife 
(Taylor, 1995). It did not at first surprise me to note that the most frequently 
occurring element in Fife parish-names was cill: Fettykil (the old name for 
Leslie parish), Kilconquhar, Kilgour (the old name for Falkland parish), 
Kilmany, Kilrennie, Kinglassie and Methil (3). It did not surprise me because 
cill is by far the most common element in Scottish parish-names, to be found 
in over 100 of them; while in Ireland it is even more frequent. However, 
Nicolaisen's distribution map ofplace-names containing this element throughout 
Scotland as a whole (for which see Fig. 5 . 1 ) clearly showed that Fife was in fact 
exceptional, since it includes one of only two clusters of cz'Z/-names which reach 
the North Sea. The other is in Easter Ross and east Sutherland. 

The next step was to verify this distribution in eastern Scotland. The 
provisional results can be seen on Fig. 5.2. Although not fundamentally 
different from Fig. 5 . 1 , several points should be noted: 

1 . In East Fife there are an additional six Ci7/-names, which brings the total tally 
to eleven (see Appendix 2.1). 

2. The three dots in south-east Perthshire disappear, leaving the east Fife cluster 
even more isolated. 


3. The two dots near Logierait on the Tay in Atholl have multiplied to five. 

4. Angus is no longer a totally cz'ZZ-free zone, with two examples beside Brechin 

Fig. 5.2 shows only those places which contain cill beyond reasonable 
doubt. It is important to stress this point, since cill is a notoriously difficult 
element to identify. It is often confused with Gaelic coille 'wood' , with cinn 'at 
the end of, and sometimes also with cuil 'corner, nook'-; or cul 'back'. For 
example, not only can cill endup as cinn, as inKinglassie FIF, but cinn can end 
up as cill. A good, if untypical, example of this change, is one that took place 
when Gaelic was still the main language of east Fife: the Gaelic name for St 
Andrews is CillRimhinn, earlier* Cz'ZZ RighMhonaidh, anglicised as Kilrymonth, 
andmeaning 'church oftheroyal hill'. It first appears as suchinthe early twelfth 
century (e.g. St A. Lib. 124). If, as is the case for 99% of Scottish place-names, 
we had no record of this name prior to 1 100, we would certainly have classed 
it, too, as a cz'ZZ-name. Exceptionally, there are earlier forms going back to the 
eighth century, and from these it is clear that it was originally * Cinn rig monaid, 
'at the end of the royal hill'. 

Yet another element which often appears as Kil- in place-names is the Scots 
word kiln. It is, however, easier to spot, since the second element is also Scots, 
and the stress pattern is usually different, falling on the first syllable. It is found 
in such combinations as Kilburn, Kilfiat, Kilknowe and Kilfield. 

If there were added to Fig. 5 .2 doubtful cz'ZZ-names, then the only significant 
change would be in Aberdeenshire, where there is a handful of problematic 
names suchasKennethmont(parish),Kilblean(Meldruiriparish) andKinbattock 
(the old name of Towie parish). But the very rarity of this element in the North- 
East is itself an argument against these problematic names containing cill. 

I want now to home in on Fife, or rather east central Scotland. In contrast to 
east Fife, there is a remarkable dearth of cz'ZZ-names throughout west Fife, 
Kinross-shire and Clackmannanshire, in short throughout most of the region or 
province which was known as Fothrif, and which gave its name to one of the 
deaneries of St Andrews (5). It is therefore of particular significance that in this 
area two medieval parishes contain the element both. These are Tullibole KNR, 
■ formerly FIF, and Tullibody CLA. Also there was a chapel of Bath, earlier 


T) n .t?1ca 

Both, near Culross FIF, formerly a detached part ^^^tZ 
lands opposite there was theoldparishoffiotoemarmSTL.(FormoredetoUs 

names benorth the Forth is that they all had links with Cuhoss «to 
proprietorial, with the Cistercian abbey there, or with events m the Life of St 
Serf, whose centre was of course at Culross. 

This is not the only part of Scotland where both occurs in an ecclesiastical 
con^tTaLgparisinames first: in what had once been the northern part of 

are four parishes which contain this element, for winch see Appendix 1.1. 

North of the Forth, in Pictish territory, there are at least seven medial 
parishes with this element. These are also listed in Appendix 12, and their 
pansnes wuu distribution is shown on Fig. 

, 1 5.3. A glance at these lists 

shows just how well 
disguised some of these both- 
elements are. This goes some 
way towards explaining why 
they have hitherto eluded 

Included in Appendix 1 .2 
are three oof/i-names which 
deserve to be mentioned in 
this discussion, but which do 
not fit into either of the ab ove 
categories. Two are medieval 
parishes lying outwith 
Strathclyde, Lennox and 
Pictland: these are Bothan 
ELO, now the parish of 
Yester; andBothkennar STL, 
which has already been 
alluded to . As it lies just south 
of the Forth, opposite 
Kincardine by Culross, itmay 

• Name containing cill 

x Name containing both (with religious connotations) 

Fig. 5.3 Map of place-names containing cUl and both 

(with religious associations) (S. Taylor) 

9 i Britain, 

be seen as an outlier of the cluster of ecclesiastical both-nzmes on the other side 
of the river. It contains a female personal name, Cainer, borne by at least two 
early Irish saints, found also in the parish-name Khkinner WIG (see Watson, 
1926, 166, 275 and 429). 

The third such name is the now obsolete Shambothy, 'the place of the old 
both'. This lay beside Clackmannan CLA. It has no specifically religious 
connotations, but it was the caput of a medieval barony, an unusual role for an 
auld sheiling! 

Both also occurs at least three times north of the Forth in conjunction with 
early chapel-sites, for which see also Appendix 1.3. 

Two ecclesiastical oo^z-names are linked with a St Ernoc, a diminutive of 
Ernene. One, combined with the second person singular possessive pronoun 
do, is the parish of Baldernock STL (see Appendix 1.1); the other, combined 
with the firstpersonpossessive pronoun mo, is *Bothmernock ANG. This gives 
a new twist to the fact that a parish bearing his name, the parish of Marnoch 
BNF, lies only a few miles east of the remarkable cluster of octf/z-parishes in 
BNF and ABD (Boharm, Botarie and Botriphnie). 

The most famous of the many saints called Ernene or Ernoc was Ernene mac 
Cresene, a younger contemporary of Columba, described by Adomnan as 
famous among all the churches of Ireland, and very widely known (Anderson, 
1961, §15b). But as Watson suggests, there might well have been a Scottish 
saint of that name not recorded by Irish writers (Watson, 1926, 292). 

-i- ^" 

The Welsh cognate bod means simply 'dwelling, residence', and is mainly 
found in place-names in north-west Wales. I have counted 24 in all, about one 
third of which contain personal names. Five of these are parishes (listed in 
Appendix 1 .4), with at least two containing the name of the saint to whom the 
parish church is dedicated (Bodedern and Bodwrog, both in Anglesey). 

Bod is also found unevenly distributed throughout Cornwall. Oliver Padel 
has counted about 230 .Bod-names, and concludes that it seems to have meant 
a dwelling more humble than tre, the standard word for a farmstead. He does 

DavljA e 
27 Britain 

not note any religious context (Padel, 1985, s.v.). 

ome Msh Language (Dublin 1913- ), both meant 'hut, bothy' or cabin , and 

Z^~T:Z ZZTcl^ Tyrone, was allegedly 
fonndedby Pa*i k and contains as its second element «»J , a very ear y 
W fox ? ctach which seems to have fallen out of use by the late seventh 

names which are attached to saints' names or chapel sites. 

So although in both Wales and Ireland there is sporadic evidence for the 
this aspect of it more fully than anywhere else m the British Isles. 

In Fife and Fothrif, therefore, there are two mutually exclusive clusters of 

. east coast of Scotland. 

With respect to east Fife names containing cill, it would appear that they 
wel mdiXg out from some important early church establishment with 
SSh connections and high political standing. The most obviou 


T„ 711 the Annals of Ulster record the death of Tuathalan abbot of 

;c TriA noint to a strong link either with Ireland or Dal Riata. lviarjone 
l^on su g sts that the evidence, slight as it is, indicates a Bangor 
pToven-ce for this entry (Anderson, 1974, 2). The place ^ *Z jti 
Sctish name and, since the third element monadh is not an Irish word (see 
w It ^26^91 ff) I would posit a form such as *Pennrigmonad. That 

ST^ve" few early place-names in Scotland which certainly contam the 

9 J Britain. 

word for 'king'. There is enough evidence from various sources for general 
agreement that Kinrimonth was also an important early ecclesiastical centre. It 
stood under the special patronage of the Pictish kings on account of its alleged 
possession of relics of the apostle Andrew. Opinions are divided as to which 
king was responsible for Kinrimonth' s elevation of religious status, but 
whoever it was, there was certainly a monastery here in the mid eighth century 
important enough for the death of its head to be noted in the Irish annals (see 
Taylor, 1995, 9 ff). 

As far as the creation of cz7/-names in Fife is concerned, we are also looking 
at a period well before the widespread Gaelicisation of eastern Scotland in the 
ninth century. The general absence of cz'/Z-names in Angus and the North-East 
must mean that cill had ceased to be a productive place-name element in the east 
by the time the Scots settled Pictland to any significant extent. Cill is therefore 
the earliest dateable Gaelic place-name element we have from eastern Scotland, 
and must date to a period before about 800 (Nicolaisen, 1976, 142-4). 

Another way of assessing not only the date, but also the wider historical 
framework of our cz7/-names is to look at the saints commemorated in them (see 
Appendix 2). I assume that Ethernan is contained in Kilrenny (Kilrethni 
c.1170) in some mangled or hypocoristic form, since he was obviously an 
important figure in the area. Kilrenny is a coastal parish, and directly offshore 
lies the Isle of May, with Ethernan as its patron saint. The name, incidentally, 
would seem to have been preserved in the place-name Aithernie in Scoonie 
parish by Leven, further down the coast; while in the old kirkyard of Scoonie 
itself a Class II Pictish stone was found with an ogham inscription which reads 
'Eddarrnonn' (ECMS vol.2, 347). The Irish Annals tell us that Itharnan and 
Corindu died amongst the Picts in 669, but we have no way of knowing whether 
this was the same person as the Fife saint (ES i 180). 

Kilminning lies on the coast just east of Crail. The saint commemorated in 
the name may have been Moinenn of Clonfert, who died in the 570s (see 
Watson, 1 926, 294-5 and 328-9). Whoever he was, his cult persisted along the 
coast of the East Neuk of Fife for many hundreds of years, and the important 
fourteenth-century royal foundation at Inverie was dedicated to him. The place 
later became known as Inverie of St Monan, and is now known simply as St 
Monance, locally St Minnans. 

Of particular relevance to this discussion is the presence of a St Moinenn 

S Britain. 

amidst the other east coast emster of d/Z-names mEaster Ross. 
andearlymodern documentation weloiowthattherewasachapelofStMonan 
somewhere near Rosemarkie (MacDowall, 1963, 12). Tins 1S a small but 
significant link between these two areas. 

The most noteworthy saint amongst the Fife dZZ-names is that of Duncan, 
or Dunchad. This is not because he was such a well-known saint, but because 
he is commemorated twice, which must be significant. Who was he? There is 
only one saint of this name mentioned in the early Irish Calendars. He is 
Dunchad abbot of lona, relative of Adomnan, who died 717, and under whose 
auspices Roman usages were introduced into that abbey (ES 1 217). 

Dunchadmaybethekey to our understanding of this ^cluster in east Fife 
He lived at a time when controversy was rife in the Scottish and Prcttsh 
churches. It was a time when the Pictish king Naiton (or Nechtan) was 
consciously turning his back on the conservative Columban church, even 
expelling them from his kingdom, and was looking elsewhere, for example to 
Northumbria, for guidance in the new ways. 

The hypothetical historical frame-work which I would like to propose to 
account for the distribution of cill and both in Fife and the adjacent area to the 
west as shown on Fig. 5.3, is this: the eft-names in east Fife were created m 
the early eighth century, spreading out from an Irish or Dal Riatan monastic 

and promoting Roman usages in accordance with the wishes of King ; Naiton. 
We know from Bede that Naiton had turned to the recent enemies of Pictland, 
the Northumbrians, for help in introducing Roman Easter and the Roman 
tonsure into his kingdom. It is therefore equally likely that he would have 
enlisted the help of the less conservative wing of the Irish or Columban church 
for the same ends. The feet tot he d^d not mtroduceNorthumbn^ clergy in o 
Kmrimonth is scarcely surprising, given that the province of Fife had only 
recently been recovered from Northumbrian domination, after the battle of 
Dunnichenin 685. 

Even more politically sensitive would have been Fothrif or west Fife and 
adjacent territories. Only a short stretch of water separated these from 
Northumbrian-held Lothian, and from Abercorn WLO, the centre of a short- 

1 °° Brimm 

lived bishopric set up by King Egfrith for his Pictish conquests (see Bede 
Historia Ecclesiastica IV.26). 

In a recent re-assessment of the Latin Life of St Serf, Alan Macquarrie 
confirms Serfs credentials as a Pictish saint, and tentatively suggests that he 
filled the vacuum north of the Forth left by the flight of the Anglian bishop 
Trumwine from his seat at Abercorn after 685 (Macquarrie, 1993, passim, 
especially 133). The church in this area would seem to be under direct Pictish 
control, which explains both the absence of ez7/-names, in stark contrast to east 
Fife, and the presence of several places containing what I hope I have been able 
to show is a Pictish and (north) Curnbric word for church, both or bod. 

Macquarrie goes on to suggest that in filling this vacuum along the shores 
of the Forth, Serf was at the same time 'reorganising the church ... along the 
lines of the Gaelic church' (Macquarrie, 1993, 133). It is more likely that, if Serf 
was indeed re-establishing the church north of the Forth after 685, he would 
have organised it along Roman and Pictish lines, under the close scrutiny of the 
Pictish kings. 

Serf s Life opens with a long section associating him with lands around the 
Mediterranean. This included a spell as pope, before, through divine intervention, 
he decides to swap Rome for Culross. What we maybe seeing here is some kind 
of symbolic portrayal of Serfs allegiance to a Roman party within the late 
seventh- or early eighth-century church. 

* * * * 

I would now like to widen out the focus again to include all of eastern 
Scotland, and to ask the question: do other cill and both names throughout 
formerPictland support any of the above hypothesis regarding their distribution 
in Fife and Fothrif? 

The five «7/-names in Atholl are all within a few miles of each other, and are 
all clustered around the important secular and church centre of Logierait, at the 
junction of the Tay and the Tunimel. Beyond this tight knot are another three 
or four in neighbouring areas both to the east and the west. Most of these are 
listed in Appendix 2.2. The earliest forms of Logierait are Login Mahedd etc. 


(e.g. RRS ii no.336, dated 1 189x95). It almost certainly contains the name of 
St Coeddi, bishop of Iona, who died in 712, and was a signatory of Cain 
Adamndin, otherwise known as the Law of Innocents (W arson, 1 926, 3 14). As 
with Dunchad above, this is another link with Iona, and Adomnan, around 700. 

Logie Mahedd was the chief church of Atholl, situated at Rait, which was 
the caput of Atholl (Scone Liber no.55). Logie Mahedd's association with at 
least some of these surrounding ci/Z-names is not purely one of geographical 
proximity. A late twelfth-century charter, by which Earl Malcolm of Atholl 
grants the church of Logie Mahedd to Scone Abbey, lists three of the nearby 
cZZZ-names as chapels belonging to Logie (RRS ii no.336). 

Noteworthy also is the place-name Ardeonaig on the south shore of Loch 
Tay, with its associated church of Cill-mo-Charmaig. This was the name of a 
medieval parish, and was originally Ardoueny (1275, Bagimond's Roll 73), 
later Ardewnan (RMS ii no.2235), and contains the name Adomnan. Douglas 
MacLean suggests that the saint' s name in Cill-mo-Charmaig may in fact be a 
hypocoristic formof Columba himself (MacLean, 1983, 60). I would not want 
to press this too far, as hypocorism can cover a multitude of sins. What is more 
certain is that Ardeonaig contains the name of Adomnan. Nor is it the only 
place-name to contain 'Adomnan' in this area. Watson (1 926, 270-7 1) lists two 
in neighbouring Glen Lyon, and one by Grantully in Strath Tay; while the 
important early church centre of Dull, also in Strath Tay, had Adomnan as 
patron saint. 

All this, taken along with various later traditions associating Adomnan with 
the area, does suggest the active presence, if not of Adomnan himself, then at 
the very least of his close associates or devotees at the heart of Atholl. It is, I 
would contend, no coincidence that it is in this very area where, in contrast to 
other parts ofPerthshire and central eastern Scotland, we have such aproliferation 
of the czZZ-element. 

A detailed analysis of the cluster of cZZZ-names around Rosemarkie in Easter 
Ross and south-east Sutherland has still to be undertaken, although it is 
tempting to see them in conjunction with that famous, if shadowy, figure 
Curadan-Boniface, bishop of Rosemarkie, contemporary of Adomnan, and 
another signatory of Adomnan' s Law of Innocents. It would seem that, like 
Adomnan himself, he was a Romaniser. For more on Curadan, see Watson, 


■9? Bct&L- 


1926, 315; and MacDonald, \992,passim. 

In summary, my hypothesis is that the establishment of these three clusters 
of cZZZ-names in eastern Scotland was a result of Columban and/or Irish 
intervention in support of the Romanising tendencies within the Pictish 
kingdom in the late seventh and early eighth century. The presence of the place- 
name element both in an ecclesiastical context points, however, more towards 
direct Pictish or British spheres of influence, possibly within the same historical 
context as cill. 

Nevertheless, in the case of the eastern czZZ-clusters, whatever they might 
have in common from an onomastic or historical point of view, there is one 
important factor which separates the east Fife cluster from those of Atholl and 
Easter Ross. That is accessibility from Dal Riata. Atholl lies at the eastern end 
of one of the few major crossing points of Druim Alban, by Strathfillan and 
Glen Dochart to Loch Tay. The name 'Atholl' alone, which means New 
Ireland, and is first recorded in 739 (see Watson, 1 926, 228-9), suggests major 
early settlement from the west, long before the general establishment of 
Scottish power in Pictland in the ninth century. Furthermore the Amra 
Choluimb Chille, the famous elegy to Columba written within a few years of 
his death, twice mentions Columba as working amongst tribes of the Tay (6). 

Easter Ross lies at the eastern end of the Great Glen, the chief link between 
Dal Riata and east northern Scotland, and the route we know Columba took on 
several occasions (see Anderson, 1961, §§74b-75b, 79b-82a and 1 14b-l 15a). 
There is in fact a trail of czZZ-names which leads from Dal Riata to Easter Ross 
along this important thoroughfare. 

So both these areas - Atholl and Easter Ross - could well have been open to 
important Columban influence during the life-time of Columba himself, over 
a century before 700. This very early contact may have played its part in the 
creation of the unusually high number of cz/Z-names in these two areas. So 
whatever else this means, it does make the cluster of czZZ-names in east Fife even 
more exceptional. 

In any final analysis, which I am in no position to attempt here, we would 
have to take into consideration other features which make Fife special in the 
Pictish period, the most strildng being the marked lack of Class I and Class II 



Pictish stones relative to other parts of Pictland (7). 

I referred above to the rich seam of ecclesiastical place-name elements m 
Scotland, and the brief glimpse which this discussion provides shows justhow 
richaseamttis^veninarelativelyunrefinedform. There is rnuchmorework 
tobedon e ,however,beforepl a ce-namescanmaketheirmllcontnbutiont 0 our 

church in Scotland. 

St Andrews Scottish Studies Institute 
University of St Andrews 

I am grateful to Dr Thomas Clancy for his comments on this text. 

Appendix 1 ** ~' ~ 

Both/bod in ecclesiastical contexts 


1 Parishes containing bod or both in Strathclyde and Lennox. 
BALDERNOCk STL ([BJuthernok & Buthirnok c.1200; Bothornok 1531) 
BALFRON STL (Buthbren 1233) 

BONHILL DNB (Buthelulle c.1270; Bothlul 1273) 
BOTHWELL LAN (Botheuill 1242) 

2 Parishes containing both in Pictland. 
BALQUHIDDER PER (Bu[t]hfyder c. 1268) 

BETHELTStlE now Meldrumpar. ABD {Buthelny c.1220; Bothelny 1452) 
BOHARM BNF originally a chapel of Arndilly (Bocharnye 1426) 
BOTARTE now Cairnie par. ABN (Butharryn 1232) 

BOTRIPHNIE BNF (Buthrothyn & Buttruthin 1226) ^ 
TULLIBODY CLA (Dunbodeuin 1147; Tulibodeuin 1164; Tuhbofthjen 

TULLIBOLE KNR (Tulybothwyn 1217; Tulibotheuile & Tulibothwin 1227) 
Note also 

BOTHAN par. now Yester par. ELO (Bofftara c.1250) 

»°* Britain. 

BOTHKENNAR par. now in Grangemouth par. STL (Buthkenner c. 1250) 
SHAMBOTHY f CLA (Sc(h)embothy c.1350, Schandbody 1490) 

3 Chapel sites etc. containing 6of/z in Pictland. 

BATH Culross par. FEF (formerly PER detached): (Baith Estir & Westir 1540; 
WestirBoth 1543; Chapeltoun of Both 1587). 

BOATH Carmylie par. ANG: the chapel of St Laurence situated on the land of 
Konanmorcapil (part of the lands of Conon) is described as the chapel "del 
Both'" 1276 x 88 (Arb. Lib. i no.247). 

BOITH j, Panbride par. ANG: the chapel of Boith was granted to church of 
Brechin, by Bishop Adam of Brechin in 1348; it was stipulated with this grant 
that vicar of adjacent Monikie par. was to say a mass of St Mernoc ('missam 
de beato Marnoco') every Sunday (Brech. Reg. i no.8); with this chapel went 
land of Botmernok in the feudal holding of Panmuir, Panbride par., 'the both 
of St Mernoc' {Brech. Reg. ino.ll). 

BORENICH, Blair Atholl par. PER: occurs with variation in the generic 
between both and mdine (?) as Montrainyche alias Disart 1509; Borannych 
1512. This alias implies an early religious association (Gaelic diseart 

Note also 'capella de Branboth' in Glen Lyon PER St A. Lib. 295-6 1214x29. 
See also Cowan, 1967, under Killinlynar. Watson (1926, 312), however, 
derives this from a genitive of the personal name Branub. 


4 Parishes containing bod 
BODEDERN Angelsey (Edern son of Nudd) 
BODFARI Flintshire 

BODWROG Angelsey (Twrog) 


5 Parishes containing both 

BODONEY Co. Tyrone {domhnach; see above p.98) 
BOHO Co. Fermanagh (plural botha) 
BOHOLA Co. Mayo (St Tola) 
RAPHOE Co. Donegal (rath + both) 


SHANBOGH Co. Kilkenny (seann 'old'). 

1 «7/-names in Fife 

FETTYKILL (now Leslie par.) (Fithldl c. 1 175) - 'wood kirk' (Gaelic fiodh). 
KILCONQUHAR par. (Kilcunkath c. 1200)- Dunchad (Duncan). 
KILDUNCAN, Kingsbarns par. (Kyldonquhane 1382) - Dunchad (Duncan). 
KILGOUR (now Falkland par.) (Kilgouerin 1224) - ? burn-name. 
KILMANY par. (Kilmannin 1202) - ? 
KILMARON, Cupar par. (Kilmaron c.1220) - Ron. 
KILMINNTNG, Crail par. (Kilmonane 1452) - Moinenn. 
KILRENNY par. (Kilrethni c. 1 170) - ? Ethernan (of the Isle of May). 
KINGLASSIE par. (Kilglessin c.1155) - Gaelic glais 'burn, stream'. 
KINGLASSIE, St Andrews & St Leonards par. (Kynglessyn c.1220) - ditto 
METHIL (medieval) par. (Methltil 1207) - 'mid(dle) kirk' (Gaelic meadhon). 

2 Some cz7/-names in Atholl PER. 
FORTINGALL par. (Forterkil 1 2 14x49) 

KILLffiCHANGIE, Logierait par. (capella deKelchemi c.1190) - Coemhi. 
KILLIECHAS SIE, Weem par. {capella de Kelkassin c. 1 190) - Cassian. 
KTLLICHONAN, medieval par. aka Rannoch, now in Fortingall PER - 

KILMAVEONAIG, medieval par., now Blair Atholl par. (Kilmeuenoc 1275)- 
? Beoghna, second abbot of Bangor, died 606 (Watson, 1926, 310). 
KILMICHEL of Tulliemet, Logierait par. (capella de Kelmichel de T.cAl 90)- 

KILMICHEL, now Kirkmichael par. (Kylmichel 13th c.) - Michael. 
KILMORICH, Dunkeld & Dowally par. - ? Muireadhach. 

Note also: LOGIERAIT, formerly Login Mahedd c. 1 190 - Coeddi. 



County Abbreviations 
ABD Aberdeenshire 
ANG Angus 
BNF Banffshire 
CLA Claclonannanshire 
DNB Dunbartonshire 
ELO East Lothian 
FIF Fife 

LAN Lanarkshire 
KNR Kinross-shire 
PER Perthshire 
STL Stirlinghshire 
WLO West Lothian 

Other Abbreviations 

Arb. Lib. Liber S. Thome de Aberbrothoc, Bannatyne Club 1848- 


Baghnond's Roll SHS Misc. vi, pp.3-77, ed. A.I. Dunlop 1939. 

Brech. Reg. Registrum Episcopatus Brechinensis, Bannatyne Club 


Calendar of Scottish Supplications to Rome, vol. 4 ( 1 43 3 - 
47), ed. A.I. Dunlop & D. MacLauchlan 1983. 

Discovery and Excavation in Scotland. 

Early Christian Monuments of Scotland, J. Romilly Allen 
& J. Anderson, 1903 (reprinted in 2 volumes, Pinkfoot 
Press, 1993). 

Early Sources of Scottish History 500 to 1286, ed. A. O. 
Anderson 1922. 


Registrum Magni Sigilli Regum Scottorum edd. I.M. 
Thomson & others 1882-1914. 

™ Britain, 

CSSR iv 




EMS ii 

Regesta Regum Scottorum vol.ii, (Acts of William I) ed. 
G.W.S. Barrow 1971. 

Scone Liber Liber Ecclesie de Scon, Bannatyne & Maitland Clubs 


St A. Lib. Liber Cartarum Prior atus Sancti Andree in Scotia, 

Bannatyne Club 1841. 

1 For the most comprehensive list of Celtic ecclesiastical place-name 
elements in Scotland, see Watson, 1926, 244-69. 

2 See for example A. MacDonald, "Aimaf in Scotland: A Provisional 
Review', Scottish Studies vol. 17 (1973), 135-46; T.O. Clancy, Annat in 
Scotland and the origrns of the parish', Innes Review, forthcoming; A. 
MacDonald, 'Gaelic cill (Kil(l)-) in Scottish Place-Names Bu lletin of the 

kirk); and 128 ff (for cill); G.W.S. Barrow, 'The Childhood of Scottish 
Christianity: aNote on Some Place-Name Evidence', Scottish Studies vol 27 
( 1983) 1-15 (for *egles); J. MacQueen, 'The Gaelic Speakers of Galloway and 
Carrick', Scottish Studies vol. 17 (1973), 17-33 (for cill and far/cm Galloway). 

3 For more details of these names, see Appendix 2.1. See also Taylor, 1995. 

4 Thesearea)Kilmuir(a7/Mofr e 'churchofMary'),amedievalparish,now 
part of Brechin par. ANG (Kilmor 1274 Bagimond's Roll 52; mistranscribed 
as Kyrimur 1275 ibid. 69). In 1473 it is referred to as the church of the Blessed 
Virgin Mary of Kilmore next to Brechin. The church was situated between the 
ISedral and the castle (National Grid Reference N0597 600). Note that the 
only carving of the motif of the Virgin and Child in early medieval Scotland, 
ap artfromIona,andKlldalton,Isla y ,comesfromaClassmstonefromBrechm 

fSmvth 1984, 126 and ECMS vol.2, 249-50). 

h) Burghill, a medieval parish, now part of Brechin par. ANG (chuixh of 

5 OMreg z7Z1473i?M5iino.ll02).Itis almost certainly the P a f hof S tTe ; a ^ 
0 frj„^i e /a7whichappearsasaprebendmthechurchofBrechinml446(C5S7? 

iv no 1290; see also Cowan, 1967, under Unthiekil). The first element is 

1 2? Britdxn 

probably Gaelic bothar, Old Irish bothar, 'road, lane'. 

5 For more on Fothrif, its extent and usage, see Taylor, 1995, 20-7. 

6 See Clancy and Markus, 1995, 104-5 and 112-3; for discussion see 118-9. 

7 See, for example, the distribution map in Stevenson, 1955, 100. To this must 
be added three more Class I stones, from the parishes of Collessie, Falkland and 
Strathmiglo respectively. The two shown on Stevenson's map are from 
Lindores, Abdie par., and Walton, Cults par. . All five lie within the same swathe 
of country north or north-east of the Lomonds in parishes on either side of the 
boundary between the medieval deaneries of Fife andFothrif. Seei?CMS , vol.2, 
343-4; Jackson, 1989, 32-3; and DES, 1989, 17. 

References ~~~~~~ ~~~~~~~~~~ 

Anderson, A.O. and M.O., 1961, Adomnan '$ Life of Columba. 

Anderson, M.O., 1974, 'St Andrews before AlexanderF, TheScottish Tradition 
ed. G.W.S. Barrow, 1-13. 

Clancy, T.O. and Markus, G., 1995, Iona: The Earliest Poetry of a Celtic 

Cowan, LB., 1967, The Parishes of Medieval Scotland, Scottish Record 
Society, vol.93. 

Flanagan, D., 1982, 'A Summary Guide to the more commonly attested 
ecclesiastical elements in place-names [with special reference to Northern 
Ireland]', Bulletin of the Ulster Place-Name Society, series 2 vol.4, 69-75. 

Jackson, A., 1989, ThePictish Trail. 

Joyce, P.W., 1869, The Origin and History of Irish Names of Places, vol. 1. 

MacDonald, A., 1 992, Curaddn, Boniface and the early church ofRosemarlcie. 

*9 Britain 

MacDowall, C.G., 1963, The Chanonry of Ross. 
MacLean, D., 1983, Scottish Studies vol.27, 53-65 

Macquarrie, A., 1 993, 'Vita Sancti Servani: The Life of St Serf , Innes Review 
vol.44 (2) 122-52. 

Nicolaisen, W.F.H., 1976, Scottish Place-Names (second impression with 
additional information 1979). 

Padel, O.J., 1985, Cornish Place-Name Elements. 
Smyth, A.P., 1984, Warlords and Holy Men. 

Stevenson, R.B.K., 1955, 'Pictish Art', The Problem of the Picts ed. F.T. 
Wainwright, 97-128. 

Taylor, R.S., 1995, 'Settlement-names in Fife', unpublished PhD, University 
of Edinburgh. 

Watson, W.J., 1926, The History of the Celtic Place-Names of Scotland. 




Iona, Scotland, and tke Cell De 

Tkomas Owen, Clancy 

Although recent scholarship has illuminated a great deal about the crucial 
period in Scottish history from around 800 to 950, there remains much work to 
be done (Anderson 1980; Smyth 1984; Broun 1994). This paper aims to 
investigate two aspects of the evolution of the church in Scotland during this 
period. First of all, I wish to address the question of the history of Iona and its 
abbots in the years following the building ofKells. Second, I willbe investigating 
the widespread presence in Scotland in the later middle ages of the groups of 
clerics known as celiDe, 'clients of God', by first asking questions about their 
origins. I will begin by looking at the career of one person, a neglected abbot 
of Iona about whom some fairly certain things can be said, and then proceed to 
some much more speculative arguments. My approach to some of the Scottish 
sources, as will become apparent, is optimistic: I take some material such as 
king-lists and chronicles as being more or less reliable. In some senses, then, 
my analysis later in the paper will be based on a best-case scenario for the 
interpretation of the sources. 

A mythology has grown up in scholarly accounts about the effects of the 
early Viking raids on Iona, which assumes that, having built the new Irish 
monastery ofKells between 807 and 814, Iona was effectually abandoned, 
leaving a hole in relationships between the Columban/amzYza and royal power 
in Scotland only rectified in 849 with the transfer of some of Columba's relics 
to the new church of Dunkeld built by Cinaed mac Alpin (for instance, Smyth 
1984, 147, 185-8). This cannotbesupportedbytheevidencewehavefromlrish 
annals. The view that leaders of the Columban community abandoned Iona for 
Kells in 814 has been recently rejected by both Maire Herbert and John 
Bannerman (Herbert 1988, 70-3; Bannerman 1993, 29-33), but there is room 
for further exploration of the role of Iona in Irish and Scottish politics in the first 
half of the ninth century. 

Alongside the gap in our understanding of religious history at this period left 
by the story of the semi-abandonment of Iona, we may place the mysterious 
success in Scotland of the religious reform movement associated with the name 
of the celi De. This movement is primarily associated with the south of Ireland, 
but by 943 we find Constantin son of Aed, king of Scots, retiring to become 


abbot of the celi De community in Cennrigmonaid (Kinrimond, later St 
Andrews) and by the twelfth century we become aware-through charters and 
other church records-of the astonishing prevalence of celiDe co~ taes 
throughout Scotland (ES i, 446-8; Anderson 1974, 2-3; Anderson 1980 283, 
Reeves 1864) We lmow next to nothing about how this reform movement was 
introduced into Scotland, let alone why its success was so pervasive here. 
Indeed, there has been little new research into the celi De in Sco land since 
Reeves producedhis study in 1864 (but see Anderson 1974, 2-3; Miller 1982, 
140-3; MacQuarrie 1992). 

This paper aims to address both these problems in Scottish history, and to 
suggestthattheinflueneeofthece/iDeinScotlandisatleastpartly connected 
with the actions of a much-neglected figure, Diarmait alumnus (fosterson or 
pupil) of Daigre, abbot of Iona from 814 to sometime after 831. 

Diarmait is an important figure and one about whom a fair amount can be 
known and hence his neglect by scholars is the more inexplicable. Only two 
causes for this occur to me. One is that he falls between two perceived 
watersheds in the history of the Columban/amffifl. The building of Kells is one 
ofmese watersheds, andDianr^tisthefirstabbotoflonaafterits completion, 
andhencedoesnotfigureinmestadyoflonaabbots in the Andersons editions 

of the Life of Columba (Anderson and Anderson 1991, xhu-xhy). So too he 
comes before the other watershed, the presumed splitting of Columba s re ics 

of Bannerman's recent study (Bannerman 1993). Falling between these stoo s 
as he does, his significance has gone unappreciated. The other possible 
explanation is his lack of both obit and genealogy: he is the only f*ff£™ 
whose date of death is unknown to us. It may be that he died outside Ireland, 
or indeed outside the British Isles. Perhaps, like his successor, he went on 
pilgrimage to Rome, though one presumes that he was not murdered en route 
by Englishmen as his successor was (See AU 854). Either way, he ceases to be 
visible after 831. Added to this, we do not even know his father s name only 
the name of his fosterfather or teacher, Daigre, about whom in turn we know 

Of his career, however, the annals give us some information (See Appendix, 
1) So we know he took over the abbacy of Iona when Cellach reined in 8 14. 

i<? Britain 

read as if it was the abbacy of Kells Diarmait took over, but this is not what it 
says (e.g. Smyth 1984, 147; this is under influence of the entry for 807 in the 
later 'Dublin Annals of Innisfallen', cf. ES i, 259). Indeed, there is little clear 
evidence that Cellach himself, the founder of Kells, was ever its abbot, let alone 
Diarmait, though it has been seen as likely that the abbacy of Kells stayed in the 
hands of abbots of Iona for the first phase of its existence (Herbert 1988, 70- 
4; Bannerman 1993, 32). As far as Diarmait is concerned, every record we 
possess referring to him, and there are many, call him abbot of Iona. 

Diarmait was abbot during a turbulent period in the history of the Columban 
familia . Violence was to be found from domestic sources as well as from Viking 
raids. The king of the Ui Neill, Aed mac Neill (Aed Oirdnide) of the Cenel 
nEogain, seems to have been involved in the killing of the head of Adomnan's 
foundation of Raphoe in 817. This must have been felt as a severe betrayal by 
the Columban familia, since Aed' s father Niall Frossach had died in pilgrimage 
on Iona in 778. That same year, 817, the community of Columba went to Tara 
to excommunicate Aed. He died two years later, and we do not know if he was 
ever reconciled with Columba's monks. Indeed, the Annals of Innisfallen tell 
us, intriguingly, that he died on a hosting in Scotland, which is where the abbot 
of Iona had gone in 818 (At, CS, but see AU 819, and Kelleher 1971, 122-3). 
This incident may have helped hasten the Columban community's shift of 
interest towards the southern Ui Neill dynasty of Clann Cholmain, and away 
from their earlier northern patrons. 

We know of two sojourns of Diarmait in Scotland. We must remember that 
this is a somewhat Hiberno-centric perspective, and had we Scottish chronicles 
his times spent in Ireland might well be regarded as 'journeys'. The first was 
in 8 1 8, and we do not know its duration. The second was from 829 to 83 1 . In 
818, Chronicon Scotorum notes that he went "with the sgrin of Colum Cille". 
In 829 we are told that he went with the insignia/relics (mini) of Columba, 
returning with them in 831. What these relics were is uncertain, though they 
may have been the bones of Columba. Columba's shrine was described as 
existing on Iona in 825 in Walahfrid Strabo's poem about Blathmac mac 
Flainn's fatal defence of it (ES i, 263-5). However, it is possible that the relics 
had travelled to Kells for the consecration of its church in 814, and that they 
were being returned to Iona in 81 8. (This was possibly for safekeeping, since 
in 8 1 8 Aed mac Neill may have seemed a more ominous threat than the Norse. 
The Annals of Innisfallen record in the same year the journey of the abbot of 

'I s BriiSun 

Louth to Munster with St Mochta's shrine in flight from Aed. See Kelleher 
1971, 122-3.) Columba's relics may have stayed in Iona, perhaps being 
removed to Kells after Blathmac' s martyrdom, and travelling with Diarmait on 
his 829-831 journey. (On terminology of shrines andscrin Choluim Chille, see 
Bannerman 1993, 18ff.) 

On the other hand, Diarmait' s two trips coincide with the reigns of two 
important kings ofbothDalRiataandthePicts, Constantin son of Fergus (790?- 
820) and his brother and successor Oengus (820-834). To each of these kings 
the foundation or building of a church is assigned in the king-lists : to Constantin, 
Dunkeld; to 6engus, Kimimond (St Andrews) (Anderson 1980, 266; see 
Appendix 2ab). This latter, if the source is believable, must denote the building 
of a church or the setting up of a new community, since we know of the 
existence of a monastery at Kinrimond in 747 (AU). While it would be 
incautious to jump to conclusions about Diarmait' s involvement in these 
alleged foundations, there is every likelihood that he was. 

The movement of the relics of Columba would not have been undertaken 
lightly, and some importantpurpose, such as the consecration of a church, must 
have led to Diarmait's journeys. Indeed, considering that Diarmait' s successor 
in 849, it is almost impossible not to connect the transport of Columba' s shrine 
in 818, with the attribution to Constantin of Dunkeld. One may note that the 
annal entries for 83 1 and 849 are very similar, and it is only as a result of the 
information in the Scottish Chronicle that we conclude that anything 
extraordinary happened in 849 (see Appendix l,2d).Itis only by inferrence (no 
doubt a correct one) that we connect the two events, and judge that the relics 
which Indrechtach brought back to Ireland were partial, some of them going to 
Cinaed's new church in Dunkeld. A similar inferrence is, I would argue, 
sensible in interpreting the conjunction of the transportation of the relics of 
Columba to Scotland by Diarmait in 818 and the founding or building of 
Dunkeld by Constantin. 

Likewise the foundation or building of a church in Kinrimond during 
6engus' reign could well be one reason for Diarmait's Scottishjourney of 829- 
831, though we may note that St Andrews has no explicit Columban 
connections and a monastery was there already in the eighth century. Simon 
Taylor has already detailed the dedications to Iona abbots and bishops which 

i* Britain, 

occur in the vicinity of St Andrews, though these probably belong to an earlier 
period (see Taylor, above). As we shall see, St Andrews' connections with the 
celi De make the association with Diarmait all the more likely. 

Already, then, the annals argue for a prominent role for Diarmait in the 
events of his time, both in Ireland and Scotland. However, it is from an entirely 
different, and untapped source, that we learn about another aspect of Diarmait' s 
career: his involvement with and influence on the monastic reforms of the celi 
De. The document usually called 'The Monastery of Tallaght' was written 
sometime in the middle of the ninth century (MT, 120-2). It is a compilation of 
the customs and sayings primarily of Mael Ruain, founder of Tallaght, and his 
disciples, but it includes traditions from other monastics who were either 
influential in or influenced by the reform. The author drew heavily on 
testimony from one of the pupils of Mael Ruain, Mael Dithruib, once a monk 
at Tallaght, and latterly at Terryglass, where he died in 841 (AFM, 840). It has 
been commented that: 

the notes have the appearance of being due to a monastic Boswell 
whose Johnson was Mael-Dithruib. The phraseology suggests that 
the compiler was not writing at Tamlachta, but perhaps at Tir-da- 
glas. (Kenney, 472) 

The author of 'The Monastery of Tallaght' had a particular interest in the 
customs of Iona monks, although his knowledge of them was perhaps second- 
hand (MT, §§47, 52, 65, 66, 68, 69, 80, 85; excerpts in Appendix 4). Prominent 
in the text is the name of Diarmait, abbot of Iona, and some of the passages make 
it clear that he was in direct contact with Terryglass monks, and hence was 
probably himself the ultimate source of the Iona information contained in the 
document. For instance we are told that Diarmait left three words with 
Carthach, bishop and abbot of Terryglass (|853) (MT §47). Another passage 
which paraphrases the words of Mael Dithruib incorporates an anecdote about 
the customs of an unnamed abbot of Iona, and it seems likely that this anecdote 
came through Mael Dithruib and ultimately from Diarmait (MT §52, and see 
§85). There is some suggestion that the document was composed after Diarmait's 
death (MT, 122). The editors of the text also suggest that it was substantially 
composed before the death of Mael Dithruib in 840. If both these hypotheses 
are correct, we could surmise that Diarmait died between 83 1 and 840. 

li? Britain 

From this document we learn of the existence in ninth-century Iona of 
something called the 'Rule of Columba', and we know of some of its contents. 
So for instance we are told: 

In Colum Cille's Rule Saturday's ration is the same as Sunday's, 
on account of the honour paid to the Sabbath in the Old Testament. 
It is only in respect of work that it is distinguished from Sunday. 
(MT §69) 

We should perhaps not think of the Rule as being a written documentor se, 
but rather the customs and usages of Iona, which may or may not have been 
committed to writing. This Rule, whatever its form, was adaptable by the abbot 
of the day. In one passage we learn the way in which Diarmait would vary the 

Colum Cille, however, kept three fasts in the year, with a half ration 
on each of them, and this half ration was liberal. As an equivalent 
of fasting, Diarmait used to allow two exactly equal rations to be 
made, whether it happened to be coarse or light food, and one of 
these to be given to God; the other he was to eat himself; and this 
serves in place of a fast. (MT §80) 

We even learn interesting details, such as that Iona kept, sometime after 
Easter, a festival called the Feast of the Ploughmen, something perhaps 
equivalent to the Jewish feast of first fruits (MT §68, and see note, 172). From 
an anecdote, too, we hear of Adomnan's intervention in a dispute over the 
abbacy of Clonmacnois, aremarkable eventiftrue (MT §85;Etchingham 1993, 

We learn that Diarmait also had connections with Clonmacnois. An 
anecdote, likely to have been told to the Terryglass source first-hand, tells how 
Diarmait and Blathmac mac Flainn (the Blathmac who was martyred in 825) 
were cautioned by one Colcu that they needed to be reconsecrated by a bishop 
after having had a man called Cu Rui die in their arms (MT §65). Colcu may 
be the Clonmacnois scholar Colcu ua Duinechda (f 796 AU) and it is probable 
that Cii Rui is the king of Cenel Loegaire, a minor midland sept of the Ui Neill 
(f 797 AU, 792 AFM). If this anecdote has any historicity, andif my identification 
of the figures in this vignette is correct, then both Diarmait and Blathmac mac 


Flainn were active in the Irish midlands in the late eighth century. 

Some sort of Rule of Columba, therefore, existed in the early ninth century, 
and mediated through Diarmait, the well-respected abbot of Iona, it was 
influential on some of the celi De reformers. Diarmait is seen in turn taking 
advice from one of them, Colcu. Blathmac, later to be martyred, was also 
remembered respectfully by the reformers. Iona, it is clear, was thoroughly 
involved in this movement of monastic reform. If this paper demonstrates 
nothing else, the existence of this potent and obvious link between Iona and the 
reformers, and between the Iona abbot and Scotland, should be clear. 

In many ways this connection b etween Iona and southern Ireland mirrors the 
earlier efforts at codifying church structure and practice which united the skills 
of two monks from Iona and the Lismore daughter-house of Dairinis in the 
compilation of the Collectio Canonum Hibernensis (Wasserschleben 1885; 
Clancy and Markus 1995, 30-1). The celiDe reform seems initially to radiate 
out from a triangle composed of Lismore, Dairinis and Daire Eidnech 
(O'Dwyer 1981, 46). It is particularly notable that Mael Ruain's teacher and 
mentor, Fer-Da-Crich, was abbot of Dairinis (t747, AU; O'Dwyer 1981, 30). 
He may well have been a monk there while the Hibernensis was being compiled 
inajomtprojectbelweenlonaandDairinis scholars. The #z'Z>ernerasz's itself was 
clearly influential on some of the literature of the celiDe (O'Dwyer 1981, 4). 

Diarmait' s involvement with the celi De gives us an extremely plausible 
basis for understanding the spread of the movement which must have occurred 
in Scotland. It is notable that the three most prominent Columban houses in 
Scotland and Ireland in the ninth century — Iona, Kells and, traditionally, 
Dunkeld— had, by the eleventh and twelfth centuries, communities of celiDe 
incorporated in them (on Dunkeld, see MacQuarrie 1992, 122-3). Though we 
know nothing about when these communities were established, it may be 
significant that Columban monasteries which later rose to prominence, such as 
Deny, do not seem to have had celi De. It may well be, then, that these 
communities were established hard on the heels of the reform movement. 
Kinrimond may have had a community by the mid-tenth century. It is worth 
considering that if the ninth-century dates are correct for Dunkeld and Kinrimond, 
then they take their place among a number of newly founded or newly 
prominent monasteries associated with the celi De, such as Tallaght and 
Roscrea. We could conjecture that whatever the nature of the ninth-century 

"i 7 Britain, 

foundations of Dunkeld and Kinrimond, communities of celiDe may have been 
established there. What would these communities have been like? 

This is not the place to go into great detail about celi De customs and 
practices (see O'Dwyer 1981; Reeves 1864; Hughes 1966, 173-93). In short, 
it is accurate enough to call this a monastic reform, dedicated to the renewal of 
the coenobitic lifestyle. Although there is, as has been noticed by scholars, a 
marked increase in the number of anchorites at this stage in the annals, it is 
somewhat less than accurate to describe the celi De as an anchoritic, or even 
simply an ascetic movement. All the documentation makes it clear that the celi 
De lived in communities and served under rules. Later notices of their 
communities, existing within larger monasteries oriented primarily towards 
pastoral care or property management or political games, make it clear that 
within this mixed development called the monastery, they were the true monks 
(See Herbert 1994, 68-9). 

But the documentation of the reform makes it clear that they were not simply 
concerned with interior reform of monastic communities or ascetic discipline 
within them. They were also deeply interested in promoting those proper 
orientations and structures for church government and pastoral care which had 
been amain concern of ecclesiastical legislators in the eighth century. (On such 
'prescriptive texts' , see Charles-Edwards 1 992.) The Prose Rule of the Celi De 
incorporates the idea of the contract for provision of pastoral care found in the 
eighth-century aic (Reeves, 21 1-15; O'Keeffe). Otherrules from 
this period seem to draw on the Hibernensis (O'Dwyer 1981, 4). The Rule of 
Mo Chutu, the celiDe stance ofwhichis clearly seen in the use ofthe frrstperson 
plural when discussing the responsibilities ofthe celi De, begins by setting out 
the responsibilities of bishops and also discusses priests and confessors (Meyer 
1 906). It is worthnoting thatthe ninth-century records of reforming monasteries 
include many bishops among their personnel, often serving conjointly as abbot, 
and MaelRuain himself was a bishop (Tallaght: 792, 812, 874; Finglas: 812, 
838, 867; Castledermot: 843, all AU). Though the cele De himself would not 
necessarily be involved in pastoral care, nevertheless, the right-ordering ofthe 
church depended on each member performing his task, and the very possibility 
of ascetic detachment demanded a peaceful and well-functioning church. 

One ofthe key themes ofthe prescriptive documents ofthe eighth century 
is the question of what 'ennobles' or 'frees' a given church (the Gaelic tennis 

i 8 britam 

soerad). The Mansteich\xs:ch.-tra.ctBretha NemedToisech sets out the church's 
rights and responsibilities thus: 

What are the good qualifications ennobling {soertho: 'which free') 
a church? It is not difficult: the shrine of a righteous man, the relics 
of saints, divine scripture, a sinless superior, devout monks; the 
seven gifts ofthe Holy Ghost, the seven grades of the church with 
their divisions and with their proper functions being in it; people 
praying for those who serve it... penitents attending the sacrifice 
under the direction of a confessor with pious sayings... 

What are the disqualifications debasing a church (doertho: 'which 
make it base')? It is not difficult: being without baptism, without 
communion, without mass, without praying for the dead, without 
preaching, without penitents... water through it onto the 
altar. . .misappropriation, private property, complaining, providing 
for clients . . .reddening it with blood, putting it under a lord. . .giving 
it as payment to a lord or a kin. (Breatnach 1989, 8-1 1) 

One ofthe key reform texts, the ninth-century Prose Rule of the Celi De, 
incorporates considerations of a like nature: 

The ennobling (soerad) of a church of God (lies in) baptism and 
communion and the singing of the intercession, with students for 
reading, with the offering of the body of Christ upon every altar. It 
is not entitled to tithes, nor to a heriot cow, nor to the third of an 
andoit, nor to the payment of a set from treasures, unless its 
reciprocal duties ofthe church are in it, with regard to baptism, and 
communion and singing of intercession for her manaig (church 
clients/'parishioners' : see Sharpe, 1 992, 1 02) both living and dead, 
and there is mass upon every altar on Sundays and solemnities, and 
there is proper furniture on each altar. No church which does not 
have its proper equipment is entitled to the full tribute ofthe church 
of God, but 'a den of thieves and robbers' is its name according to 
Christ. (Reeves 1864, 211-2, translation mine.) 

One ofthe interests of eighth-century legal reformers in Munster, in Armagh 
and in the Iona-Dairinis project ofthe Collectio Canonum Hibernensis, then, 

U? Britain 

was the question of the relationship of church and state, of what a church must 
do to deserve its 'free' or 'noble' status. Without fulfilling those obligations, 
the church would lapse into subservient status, being ruled over by other 
ecclesiastic or even secular lords. These concerns were put into action by the 
celi De in 811, when the community of Tallaght blockaded the fair of Tailtiu 
as a result of Aed's violation of their sanctuary (AU, 811). The Columban 
community under Diarmait, as we have seen, acted on the same premise against 
Aed in 817. Some years earlier, another reforming cleric, Fothad na Canoine 
of Fahan, extracted for the clergy exemption from military hostings (AU, 804; 
Hughes 1966, 191-2). The already mentioned Rule of Mo Chutu, which is also 
called the Rule of Fothad na Canoine, includes a section on the proper conduct 
of Christian kings (Meyer 1906, 314-5). 

In such a context it may prove worthwhile to look again at some of the 
activities of Scottish kings in the late ninth century with regard to the legal 
standing of the church. I would argue that, to the extent that these notices are 
reliable, they reflect the influence of eighth-century Irish church legal reform, 
promoted through the monastic reform of the celi De, quite likely as propagated 
by Iona abbots. Beyond this, some of the kings of ninth-century and tenth- 
century Scotland can be seen to have then own connections with the reform. 

We have already seen that it is probable that Diarmait' s visit to Scotland 
in 8 1 8 with Colum Cille' s relics was connected with the foundation of Dunkeld 
by Constantin son of Fergus. In this regard, there is another piece of evidence 
connecting Constantin to the celiDe reformers. In the Martyrology of Tallaght, 
composed between 828 and 833 (ORiain 1990, 3 8), under the entry for March 
1 1 , the compiler(s) felt some doubt about the identity of the St Constantin there 
commemorated, to the extent of questioning whether this might be the Pictish 
Constantin, son of Fergus (Best and Lawlor 1 93 1 , 22; see Appendix 3). So near 
to the date of his death, this can only be explained in terms of his reputation for 
holiness having spread to the community of Tallaght, and it seems sensible to 
presume that this reputation was based primarily on his foundation and 
endowment of Dunkeld. We may wonder, then, whether his commemoration 
in a Tallaght document suggests that Dunkeld, when initially founded, was a 
celiDe monastery (see also MacQuarrie 1992, 122-3). 

There is other evidence of Constantin's religious associations. The recently 
discovered inscription on the face of the Dupplin cross bears his name 
(CustantinfiliusFhcus: see Forsyth, 1995), and coupled with the iconography 

£° Britain, 

of the cross is a vibrant monument to Christian kingship. He is one of three 
Pictish kings whose name is entered in the Liber Vitae of Durham (Airlie 1 994, 
42-3), and in the so-called 'Dunkeld Litany', it is almost certainly Constantin 
who is invoked among the nomina sanctorum confessorum at monachorum, 
where, slxikingly, his name is placed next to that of Diarmait (again, almost 
certainly the Iona abbot of that name) (Forbes 1872, lx). 

An entry in the Scottish King-List, here quoted from King-List D, notes that 
Giric son of Diingal (878-889) 'was the first to give liberty to the Scottish 
church, which was in servitude up to that time, after the custom and practice of 
the Picts' (Anderson 1980, 267; Appendix 2c). Many authors, such as Isabel 
Henderson and Marjorie Anderson, have noted that Pictish royal power does 
indeed seem to have extended into the religious arena (Henderson 1 967, 82-90; 
Andersonl980, 198; 1982, 120-1, 127-8; Hudson 1994).Noteworthy evidence 
of this is the decision by Nechtan mac Derile to conform to Roman usage in 
Pictland, followed by his expulsion of Columban clergy. Added to this must be 
the overt royal foundation of churches, a habit virtually urrknown in Ireland, but 
one which in Scotland includes Abernethy andDunkeld, andperhaps Kinrimond. 
Scone and Meigle appear as 'royal' institutions, and the monastery of Brechin 
is 'given to God' by the Scottish king. The royal foundation of churches, the 
power of kings to endow and construct religious centres, continued, then, 
beyond Pictish times, and the action may have contained implicit proprietary 
rights which the new Gaelic ascendancy saw as needing reformed. 

If the entry on Giric is reliable, then, I would argue that it testifies to the need 
to reach an accommodation between church and state under law, and that the 
language is in accordance with the notion of soerad, ' freeing, ennobling' , found 
in Irish legal texts of the eighth century and the reform texts of the ninth. The 
more specific process of exempting particular churches from particular types 
of tribute or duty seen first in Ireland in 951 and in the Middle Gaelic Life of 
Adomndn, and later in the Book of Deer, does not seem applicable to this entry 
(Herbert 1988, 161-2; Jackson 1972, 91-3). Rather it seems to refer to general 
principles such as those we find the celi De defending in Ireland in the early 
ninth century. Giric may then have been setting into motion a process of reform 
of the structures of the church. 

In this context we may note that the 'Dunkeld Litany' would appear to have 
its origins during Giric 's reign. The document has been subject to later 

!£' Britain 

antiquarian interference, but on cursory analysis it does appear to have a 
genuine ninth-century core. Amid other invocations are two which are of 
interest, if they are authentic: 

Ut Episcopos, Abbates Kiledeos, et omnem populum totius Albaniae 
conserves et protegas. 

UtRegem nostrum Girich cum exercitu suo ab omnibus inimicorum 
insidiis tuearis et defendas. 

That You would preserve and protect the bishops, abbots and cell De, 
and all the people of the whole of Alba (Britain? Scotland?) 
That You would guard and defend our king Girich with his army from 
all the snares of his enemies. (Forbes 1872, lxiii) 

Equally remarkable in this context is the agreement reached between 
Constantin son of Aed and the bishop, Cellach. In 906, the Scottish Chronicle 
in the Poppleton Manuscript tells us: 

Constantin the king and Cellach the bishop, on the Hill of Faith 
near the royal monastery of Scone, swore to keep the laws and 
disciplines of the faith and the rights of the churches and the 
gospels, in the same manner as the Irish (pariter cum Scottis) . From 
that day that hill has merited its name: the Hill of Faith. (Anderson, 
1980, 251; see Appendix 2e) 

This establishment of parity between the practices of the Irish church and the 
Scottish church (for I believe that is how this passage shouldbeunderstood) can 
be read with more conviction in light of what we now know about the 
theoretical, prescriptive basis of the legal standing of churches in the eighth and 
ninth centuries in Ireland. Documents from this period over and over again 
state their interest in proper order, in churches fulfilling then contract to then- 
people, in the provision of pastoral care (Charles-Edwards 1992, 70-1; 
Etchingham 1991, 104-5). These interests were as much a part of the celi De 
reformers' agenda as was monastic life. Indeed, the structural changes in the 
Scottish church which become evident from the middle of the ninth century on 
may well reflect the views of the reformers. Suddenly, we find ourselves in a 
Scotland with visible bishops, and with Fortriu, for a time at least, having a 
primepscop, a 'chief bishop', based probably at Dunkeld (AU, 865). Other 
bishops, such as Cellach follow in the records through the tenth and eleventh 


Darl 5\ae 
'S. 2 Britain 

centuries (see Anderson 1974, 2-5). 

As already noted, the reformers ' concern with structure and pastoral care led 
at least one rule to begin with a passage describing the proper behaviour of 
bishops. Riagail Phdtraic, extensively quoted by the ProseRule of the CeliDe, 
states clearly that every chief territory/kingdom (prim-tuath) should have a 
chief bishop (prim-epscop): 

There should be a chief bishop (prim espoc) in each chief territory 
in Ireland, for ordaining priests, for consecrating churches, for 
being confessor to rulers and church-leaders and priests, for 
sanctifying and blessing their children after baptism, for ordering 
works in every church, and training boys and girls for learning and 
devotion... (Reeves 1864, 213; and see Etchingham 1994) 

As well as being monastic, the reformers were episcopal. What we see in late 
ninth- and early tenth-century Scotland may well be the fruit of celi De reform 
as introduced through the influential channels of the abbots of lona and their 
colleagues (on this, see also Miller, 1982, 140-3). 

We began with a personality, that of Diarmait, abbot of lona from 814 to 
before 840. Another large figure closes this study. We have seen how 
Constantin son of Aed was instrumental in instituting some sort of reform of the 
legal standing of the church in Scotland. Helping to confirm the view that the 
sort of reform instituted at this period stems from the vision of the celi De is the 
fact that it is with his name that the first explicit mention of celi De in Scotland 
is connected. In 943 he retired into the monastery of Khirimond/St Andrews, 
to live out his life in penitence under a monastic rule. One of the king-lists 
further adds the detail that he served as head of the celi De (ES i, 447-8). The 
royal respect and patronage this implies suggests that they in turn had been 
influential in the Scottish church during his lifetime. (On other celi De 
attributes-in St Andrews, see Anderson, 1974, 1-3; Miller, 1982, 141.) 

So whether all the connections I have attempted to make here are valid or 
not, we are left with two ends of an important period, a period which begins 
with an abbot of lona, influential in the celi De reform and a sojourner in 
Scotland, whose travels with the relics of Columba perhaps coincided with 
important new foundations there. It ends with a king, influential in church 

~ Brvt?an 

reform in Scotland, who himself retires to one of the monasteries built or 
endowed a century before, living out his life as head of one of the reformed 
coenobitic communities whose ethos the abbot of Iona had helped to mould. 
Between those two brackets there are, I believe, good grounds for interpreting 
Iona, and its dealings with the Scottish royalty, as the main vehicle for the 
establishment of cell De reforms, monastic and structural, in the Scottish 

Department of Celtic, 
University of Edinburgh 

I would like to thank John Bannerman, Gilbert Markus and Simon Taylor 
for reading drafts of this paper and for helpful discussion and criticism. I am 
indebted to Katherine Forsyth and Colman Etchingham for sending me copies 
of their work in advance of publication. This work was made possible by the 
British Academy, which supports the writer with a Post-Doctoral Research 

Documents relating to Diarmait, abbot of Iona fl. 814 x 831. 

8 1 4 (AU) Cellach, abbot of Iona, having finished the construction of the church 
of Kells, resigned the headship, and Diarmait, fosterson (or pupil) of Daigre, 
was appointed in his place. 

815 (AU) Cellach, son of Congal, abbot of Iona, fell asleep. 

8 1 7 (AU) MaelDuin son of CennFaelad, superior of Raphoe, of the community 
of Columba, was slain. 

Columba's community went to Tara to curse ('excommunicate') Aed. 

818 (CS) Diarmait, abbot oflona, went to Scotland with the reliquary of Colum 


8 1 9 (AT) Death of Aed son ofNiall, king of Tara, on a hosting in Alba. [But other 
annals note his death as occuring at Ath da Ferta in Mag Conaille.] 

825 (AU) Constantfn son of Fergus, king of Fortriu, dies. 

825 (AU) The violent death of Bla[th]mac son of Flann at the hands of the 
pagans in Iona. 

829 (AU) Diarmait, abbot oflona, went to Scotland with the insignia (or relics) 
of Columba. 

83 1 (AU) Diarmait came to Ireland with the insignia (or relics) of Colum Cille. 

849 (AU) Indrechtach, abbot of Iona, came to Ireland with the insignia (or 
relics) of Columba. 

2. Scottish Notices on Kings and Churches: 

(from Anderson 1980) 
Regnal List D: 

&.Constantinus f. Fergusane xlv annis reg. Iste edificavit Dunkeldin. 

b. Hungus f. Fergusane ix annis reg. Iste edificavit Kilremonth. 

c. Girg mac Dungal xii a. reg. et mortuus est in Dunduin et sepultus est in 
Iona insula. Hie subjugavit sibi totam Yberniam [I=Berniciam] et fere 
totam Angliam et hie primus dedit libertatem ecclesie Scoticane qui sub 
servitute erat usque ad Mud tempus ex consuetudine et more Pictorum. 

Giric, Dungal' s son, reigned for twelve years; and he died in Dundurn, and was 
buried in the island oflona. He subdued to himself all Ireland [I=Bernicia], and 
nearly all England; and he was the first to give liberty to the Scottish church, 
which was in servitude up to that time, after the custom and practice of the Picts. 

Scottish Chronicle from the Poppleton MS 

d. Septimo anno regni sui reliquias sancti Columbe ad 
ecclesiam, quam construxit. 

In the seventh year of his (Cinaed's) reign, he transported the relics of St 
Columba to a church which he built. 

e. Ac in .vi. anno Constantinus rex et Cellachus episcopus leges et 
disciplinasque fidei atque iura ecclesiarum ewangeliorumque pariter cum 
Scottis in colle credulitatis prope regali ciuitati Scoan deuoueruntcustodire; 
ab hoc die collis hoc meruit nomen id est collis credulitatis. 

*» BiatSfn, 

And in his sixth year, Constantin the king and Cellach the bishop, on the Hill 
of Faith (?) near the royal monastery of Scone, swore to keep the laws and 
disciplines of the faith and the rights of the churches and the gospels, in the same 
manner as the Scots. From that day that hill has merited its name: the Hill of 

3. Martyrologies; 

TheMartyrologyofTallaght,%2%x%33 (Best and Lawlor 1931; ORiain 1990): 
1 1 March: Constantini Briton f meic Fergusa do Cruthnechaib. 
[The feast] of Constantin the Briton (or 'of the Britons'?), or the son of Fergus, 
of the Picts. 

4. Entries related to Diarmait in Monastery of Tallaght 

§§ as in MT §§47, 52, 65, 66, 68, 69, 80, 85. 

§47. Three words Diarmaid, abbot of Iona, left with bishop Carthach [of 
Terryglass, f 851 AFM]: pittance, perseverance, vigil. That is, do not make a 
resolution, 'This is the pittance I will always eat. I will say the Beati perseveringly 
without desisting. This is the vigil I will always perform.' 

§52.In the case of penance laid on sickly persons, this is what he (i.e. Mael- 
Dithruib of Terryglass) thinks right, as to the continual preparing for meals: 
alternate reviving and mortifying is practised on them, lest the perpetual 
confinement should cause their death; and if this is done, if it can be managed, 
without their knowledge, by telling his servant privately, 'Let a seland be 
brought to them in their pottage or on bread' (but it is more usual to bring it to 
them in the pottage). Once it happened that the abbot who was inTona saw that 
the recluses had a bad colour. Thereupon he went to the cook and hhnself made 
the pottage for that day. He added one-third of water to the daily allowance and 
boiled the water. When this third had boiled away, he put a lump of butter on 
each man' s allowance, and boiled it on the water, and then put meal over it, and 
so he did every day. then they noticed the change in then colour, and knew not 
what had caused it, since they saw the usual ration unchanged. So when then 
colour came back and they revived, he continued alternately to mortify and 
revive them from their dying state after this fashion. 

§65. Now, to eat a meal with a dead man, though saintly, in the house is 
forbidden; but instead there are to be prayers and psalm singing on such 
occasions. Even one in orders who brings the sacrament to a sickman is obliged 

™ B^tam 

to go out of the house at once thereafter, that the sick man die not in his presence; 
for if he b e present in the house at the death, it would not be allowable for him 
to peform the sacrifice until a bishop should consecrate him. It happened once 
on a time to Diarmait and to Blathmac mac Flaind that it was in then hands that 
Cu Rui expired. When he died, they were about to perform the sacrifice 
thereafter, without being reconsecrated, till Colcu hindered them from doing 
so. The authority is Leviticus [Lev 21.1-2; 11-12.]; and Diarmait also, the 
Abbot of Iona, agreed with him on that occasion. 

§68. He does not commend fasting: he prefers a measured pittance. There is no 
Rule where it is imposed, except on account of injury done. There is one fast 
in Comgall's Rule - namely, the Wednesday before Easter. However, Colum 
Cille recognized three fasts only in the year: the eve of Epiphany - that is, 
twelve days after Christmas, and the eighth part of Colum Cille's loaf at that 
time, with.a.selanda.nda.bochtan ofgoodmilk: that was the manner of that fast; 
and the first Wednesday of Lent, and the first Wednesday after Pentecost: the 
eighth of a loaf to each fast. However, Colum Cille relaxed the fast of the 
Passion for the saints of Ireland, because old men died of that fast after the long 
privations of Lent. A great festivity and merrymaking was regularly allowed by 
Colum Cille thereafter to the brethren: the growth of the crops was given to 
them then: three months were spent in tending and watering them. He called 
that the Feast of the Ploughmen, because it was then that the crops reached their 
full growth. 

§69. In Colum Cille's Rule Saturday's ration is the same as Sunday's, on 
account of the honour paid to the Sabbath in the Old Testament. It is only in 
respect of work that it is distinguished from Sunday... 

§80. Now, continual fasting was not practised by Comgall, and it is not 
practised by the saints at present, save one fast, namely the eve of Maundy 
Thursday after the Wednesday. On the eve of the Passion, however, no fast is 
to be observed. Colum Cille, however, kept three fasts in the year, with a half 
ration on each of them, and this half ration was liberal. As an equivalent of 
fasting, Diarmait used to allow two exactly equal rations to be made, whether 
it happened to be coarse or light food, and one of these to be given to God; the 
other he was to eat himself; and this serves in place of a fast. 



AFM=The Annals of the Four Masters, ed. John O'Donovan, Dublin, 1 848-5 1 . 

M=The Annals of Inisf alien, ed. Sean Mac Airt, Dublin, 1951. 

AXJ=The Annals of Ulster ( to A.D. 1131), ed. Sean Mac Airt and Gearoid Mac 
Niocaill, Dublin, 1983. 

CS=Chronicon Scotorum, ed. W.M. Hennessy, London, 1866. 

ES=A.O. Anderson, Early Sources of Scottish History , 2 volumes, Edinburgh, 

MT= 'The Monastery of Tallaght' , ed. E. J. Gwynn and W.J. Purton, Pro cedings 
of the Royal Irish Academy 29C (1911) 115-179. 

Airlie, Stuart 1994, 'The View from Maastricht', in Crawford, 33-46. 

Anderson, A.O. and Anderson, M.O. 1991, Adomndn's Life of Columba, 
Oxford (revised edition). 

Anderson, M.O. 1974, 'St Andrews before Alexander V, in G.W.S. Barrow, 
ed., The Scottish Tradition, Edinburgh. 

Anderson, M.O. 1980, Kings and Kingship in Early Scotland (revised edition) 

Anderson,M.0. 1982, 'Dalriadaand the creation of the Kingdom ofthe Scots', 
in Dumville, et aL, 106-32. 

Bannerman, John 1 993 , ' Comarba Coluim Chille and the Relics of Columba' , 
Innes Review 44, 14-47. 

Best, R.I., and Lawlor, H.J., eds 1931, The Martyrology of Tallaght (Henry 
Bradshaw Society, vol.68) London. 

Dark _„ 
W Briton 

Blair, J. and Sharpe, R., eds 1992, Pastoral Care Before the ParaA, Leicester. 

Breatnach, Liam 1989, 'The fust Mid of BrethaNemedTolsech',Eriu 40, 1- 

Broun, Dauvit 1994, 'The origin of Scottish identity in its European context', 
in Crawford, 21-31. 

Charles-Edwards, T.M. 1992, 'The pastoral role of the church in the early 
Irish laws', in Blair and Sharpe, 1 10-33. 

Clancy, Thomas Owen and Markus, Gilbert 1995, Iona: the earliest poetry of 
a Celtic monastery, Edinburgh. 

Crawford, Barbara E., ed. 1994, Scotland in Dark Age Europe, St Andrews. 

Dumville, D., McKitterick, R. and Whitelock, D. 1982, Ireland in Early 
Medieval Europe, Cambridge. 

Etchingham, Cohnan 1991, 'The early Irish church: some observations on 
pastoral care and dues', Eriu 42, 139-62. 

Etchingham, Cohnan 1993, 'The implications of paruchia \Eriu44, 139-62. 

Etchingham, Colman 1994, 'Bishops in the early Irish church: are-assessment', 
Studia Hibernica (forthcoming). 

Forbes, A.P. 1872, Kalendars of Scottish Saints, Edinburgh. 

Forsyth, Katherine 1995, 'The Inscriptions on the Dupplin Cross', From the 
Isles of the North: Medieval Art in Ireland and Britain (Proceedings of the 
Third International Conference on Insular Art, Belfast, April 1994), Belfast. 

Henderson, Isabel 1967, ThePicts. London. 

Herbert, Maire 1988,/owa, Kells andDerry: The History and Hagiography of 
the Monastic Familia of Columba, Oxford. 

Herbert, Maire 1994, 'ChartermaterialfromKells',inF. O'Mahony, The Book 
of Kells: Proceedings of a conference at Trinity College Dublin, 6-9 September 
1992, Cambridge. 

Darl^A e 
»S» Britain 

Hudson, Benjamin T. 1994, 'Kings and Church in Early Scotland', Scottish 
Historical Review 73, 145-70. 

Hughes, Kathleen 1966, The Church in Early Irish Society, Ithaca, NY. 

Kelleher, John 1971, 'The Tain and the annals', Eriu 22, 107-27. 

Kenney, James F. 1929, The Sources for the Early History of Ireland: 
Ecclesiastical, New York. 

Jackson, Kenneth 1972, The Gaelic Notes in the Book of Deer, Cambridge. 

MacQuarrie, Alan 1992, 'Early Christian religious houses in Scotland: 
foundation and function', in Blair and Sharpe, 110-33. 

Meyer, Kuno 1906, 'RegulaMucutaRaithni',inW. Stokes and K. Meyer, eds, 
Archivfur celtische Lexicographic, Halle. 

Miller, Molly 1982, 'Matriliny by treaty: the Pictish foundation-legend', in 
Dumvilleetal., 133-61. 

O' Dwyer, P. 1981, Celi De: Spiritual reform in Ireland 750-900, Dublin. 

O' Keeffe, J.G. 1904, 'The Rule of Patrick', Eriu 1, 216-24. 

6 Riain, Padraig P. 1990, 'The Tallaght Martyrologies, redated', Cambridge 
Medieval Celtic Studies 20, 21-38. 

Reeves, William 1864, On the Celi-De, commonly called Culdees. (originally 
RIA Transactions xxiv). 

Sharpe, Richard 1984, 'Some problems concerning the organization of the 
church in early medieval Ireland', Peritia 3, 230-70. 

Sharpe, Richard 1992, 'Churches and communities in early medieval Ireland: 
towards a pastoral model', in Blair and Sharpe, 1 10-33. 

Smyth, Alfred 1984, Warlords andHoly Men: ScotlandAD 80-1000 , London. 

Wasserschleben, Hermann 1885, Die Irische Kanonensammlung, Lepizig. 

2° Britain 

mevcjence of the Recjnwm Scottovum: cl Carolincjian 

Patrick Wormald 

The first, central and I hope abiding message of this paper, all else that follows 
notwithstanding, is that we do not know how the kingdom of the Scots came 
into being, and we never shall. It was one of the two formative political 
developments of early medieval British history. It happened at more or less the 
same time as the other, the Making of England. Unlike the Making of England 
(though, as I shall stress, not so very unlike), it can also be called the last maj or 
development of British prehistory. It was the last time that there was a 
significant change in the political and cultural contours of the British Isles 
without our being able to say much about how, let alone why. The Emergence 
of the Regnum Scottorum is, then, an object lesson in the frustrations of life as 
an early medieval historian. At the same time, a subtext of my paper, addressed 
to any prehistorians whom it may concern, is that the current vogue for denying 
that there ever was disruptive change (apart from internal social revolution), 
until the existence of written sources obliges us to admit as much, is overdue 
for reconsideration. 1 

What do historians do when confronted with what amounts to a wall of 
silence (or, at best, a burble of distant and indistinct voices)? One thing they 
can do is what Anglo-Saxon historians did until lately with their equivalent 
problem, which is not the emergence of 'Engla Lond ' but the transformation of 
lowland Britain between the early fifth and late sixth centuries. Each shard of 
surviving evidence is seized and crammed into its place in an interpretative 
vessel so crafted as to accommodate them all. The problem with this approach 
is that the value of evidence does not, unlike that of other commodities, actually 
increase with its scarcity. Material may come to assume for historians a 
significance quite other than that which it had for its authors: for example, we 
now tend to see the pre-conversion annals of the Anglo-Saxon Chronicle not as 
a record of the conquest of southern England in the sixth century, but as some 
sort of explanation and legitimation of the way it was ruled in the ninth. The 
difficulty may be enhanced when handling the products of Celtic culture, with 
its inclination to be not so much economical as inflationary with the truth (a far 
more grievous sin in the canon of our current masters); and with its positive 

«0 Britain, 

genius for the manipulation of the regnal records which were part of its stock- 
in-trade. One of the blessings of being a Scottish early medievalist must be the 
legacy of Skene and Anderson: everything you need to know is brought 
miraculously within the compass of one binding. But it is apt to have the effect 
of reducing all sources to a common level of acceptability or otherwise; of 
playing down or overlooking the extent to which various texts may have 
varying slants - may even, if pulled apart and examined as wholes, be trying 
to say opposite things. 

Another approach is that currently favoured by prehistorians, by some 
students of the Adventus Saxonum, andnow by some interpreters of the Making 
of a Scottish kingdom: denial that anything very dramatic happened at all. The 
hints at continuity that the sources certainly offer are taken to preclude the sort 
of violent takeover that the same sources emphatically do not rule out. A 
pervasive view over the past quarter century sees the ground for Kenneth mac 
Alpin's ascendancy as prepared by two hundred and fifty years of a creeping 
' Scotticization' , above all through the impact of the Columban Church, but also 
through the gradual replacement of Pictish matrilineal regnal inheritance by 
Gaelic norms (Bannerman, 1971,79; Anderson, 1982, 108-15). 2 Ihavetosay 
right at the start that this line of thought is in my view open to one grave if not 
insuperable objection, namely the fate of the Pictish language. As I understand 
such things, Pictish was a P-Celtic tongue, so a lot more like Welsh than Gaelic 
(Forsyth, 1995a, further to Jackson, 1955, and Smyth, 1984,46-52). Itakethis 
to mean that Gaelic and Pictish were no more mutually intelligible than Welsh 
and Irish, whose membership of a single linguistic family was undetected until 
the sixteenth century. There is no question of assimilation of Pictish into 
Gaelic. Nor has much evidence of significant Pictish substrata in Gaelic (yet) 
been unearthed (but cf. Greene, 1972, 1983; Macaulay, 1975). The linguistic 
pattern seems closer to that produced by the Anglo-Saxon than the Roman, 
Norse or Norman Conquests. 

Now, prehistorians and others presently disinclined to credit the movement 
of large bodies of people across the map at any stage of the human experience 
are given to the nostrums that we do not know why or how languages change, 
and that physical removal is far from the only available explanation. This is 
true. It is also true that enforced domination, accompanied by some degree of 
migration, is a more obvious explanation than most, unless there are sound 
reasons to override it. The very latest account of the rise of the Scottish 

l # Britain, 

kingdom repudiates the taste of earlier generations of scholars for 'empire- 
building with battles ' , and goes so far as to call it ' a process that extended over 
a long period of time, (which) may have provoked so little notice because it was 
so unremarkable' (Hudson, 1994, 34-6, etc.). I fear it may symptomatize this 
interpretation that, for all that I can see, it is not once mentioned that Picts and 
• Scots spoke different languages, or that the former's would become extinct. 3 
No one today is going to elide the processes of political and linguistic change. 
No one, I presume, now supposes that Pictish was eliminated at once, or by 
'ethnic cleansing' - though the phrase reminds us, as Patrick Sims-Williams 
and Tom Shippey recently observed in the TLS, that these things do happen. 
What seems inconceivable is that it could have been eclipsed, in the short or 
long term, without displacement of the elite that spoke it. We might fairly 
wonder whether a culture whose art (at least on one reading of its symbols: 
Thomas, 1961, 1963; Henderson, 1967, 11 5-60) shows every sign of vigorous 
attachment to abroadly indigenous repertoire, is likely to have lamely abandoned 
its speech. 

As we leave that matter for the time being, there is one obvious proviso to 
beentered. Iammakingmycaseinalanguagethatwasnotimposeduponthese 
parts by force. Scotland from the twelfth century is a conspicuous example of 
a society which, mysteriously yet beyond serious cavil voluntarily, exchanged 
its ancestral tongue for that of a neighbour who was not always affectionately 
regarded. But before we are tempted to extrapolate from twelfth century 
circumstances to those of the ninth, there are two other points to bear in mind. 
For one thing, the activity of the French-speaking nobles and English hangers- 
on who provided the vehicle for lingustic and cultural change was of course 
anything but peaceful elsewhere (Davies, 1990; Frame, 1990). The context, 
that is to say, was one of violence, even if Scots were spared its manifestations 
to an extent that Welsh and Irish were not. Secondly, the twelfth century saw 
massive cultural change in the West as a whole. Professor Robert Bartlett's 
remarkable recent book (1993) illuminates as never before the ways in which 
'minority' cultures channelled into the European mainstream, at times because 
they had to, but also on occasion from choice. There is no counterpart to such 
processes in the 'Barbarian West' : among my messages in this paper is that the 
Carolingian 'Renaissance' meant not what that of the twelfth century meant for 
the Mecklenberg Slavs, to whomBartlett compares the Scots, but the experience 
of ninth century Saxons - an experience that incidentally had among its side- 
effects the early medieval West's other proverbial vanishing act, that of the 

l S? Britain 

once mighty Avars. Twenty-five years study of the Barbarian West's warrior 
aristocracies has yet to acquaint me with one that rolled over and died of 

If we cannot solve the problem of Scotland's emergence merely by piecing 
bits together or even by pretending that there is no problem to solve, what are 
we to do? The answer of course is, to guess. But we can and should control our 
guesswork by doing what three admirable papers did at the last 'Dark Age 
Scotland' symposium within these walls (Crawford, 1994): that is, to examine 
the pattern of comparable developments in parts of the West whose 
documentation is not so exiguous. Rather than read back in time from the 
known to the unknown, the method patented by Maitland' s classic Domesday 
Book and Beyond (1 897), we can read across in place from the brightly lit to the 
dimly visible. Hence, I can make clear at this point that most (not all) of what 
I mean by a 'Carolingian hegemony' is the kind of statecraft being practised 
elsewhere in the contemporary West. Can we perceive a shape to events in 
other eighth/ninth century kingdoms that would make sense of what little we 
can see of the Scots? 

The first theme that I want to pick out is precisely the apparent continuity that 
so dominates current perceptions of the subject. It is in fact the message of the 
early parts of the Table (below pp. 148-50), which aims to marshal all the main 
evidence bearing on the issue. If the eye of faith is up to penetrating the 
mysteries of material where, as Professor Duncan once remarked, 'rationality 
departs from our sources' (Duncan, 1975, 54), col. 4 will show that Dal Riatan 
tradition as reflected in eleventh century sources, and Pictish records as 
mediated by the Picto-Scottish Icing-lists in the reign of Malcolm III, were 
together maintaining that Picts and Scots had three or four kings in common 
through the half century to 839: perhaps a Conall, certainly a Constantine son 
of Fergus, an Angus son of Fergus and an Eoganan son of Angus. All these 
Icings appear in the one contemporary source, the Annals of Ulster, where the 
first is found fighting 'among the Picts' but being killed 'in Kintyre', while the 
other three are kings neither of Dal Riata nor of the Picts but 'of Fortriu' . 4 Just 
to complicate matters, however, there are contemporaneous Dal Riatan Icings 
who are not in Pictish lists: Domnall and/or Donncorci and Aed mac Boanta; 
and vice versa: Drest son of Constantine, and Talorgan (merged in the 'Q' 
tradition), thenFerat, Bridei and another trio confined to the ' Q ' list, which ends 
with a Drest killed at Forteviot or Scone. 

# Britain 

This tableau has naturally attracted a deal of skilled remoulding and 
realignment, notably from the deft hands of Marjorie Anderson and the 
vigorous grip of Benjamin Hudson. Approaches tend to fall into two lines of 
thought. The first takes the kings the two traditions share as essentially Picts 
(Skene, 1876, I, 301-16; Duncan, 1975, 54-9; and (with modifications) 
Anderson, 1982, 108-15). They may well have had Dal Riatan fathers, as 
Pictish matriliny would allow; if so, they were indeed legitimate heirs to each 
throne, so that a relatively peaceful union of Pict and Scot would be feasible in 
terms of either's values. Still, inherent in this scheme is a degree of Pictish 
domination over Scots, in that Constantine was apparently king of the former 
well before he ruled the latter. On this view, the role of Kenneth mac Alpin was 
to turn the tables on the Pictish masters, while exploiting the climate of 
subjection to a single monarch that they had fostered. The other interpretation 
(Chadwick, 1949, 127-33; Smyth, 1984, 177-85; Hudson, 1994, 29-33) plays 
down any Pictish matrilineal factor (anyway in retreat in so far as Drest and 
Eoganan at least were sons of kings), and sees Constantine and company as 
Scots. 5 They are thus harbingers of Kenneth's triumph; for Hudson, it is in fact 
they, not Kenneth, who deserve the palm of conquerors of Pictland. All that 
Kenneth did was secure their throne for his branch of the Dal Riatan dynasty; 
his family's ultimate monopoly of it was what ensured the immortality of his 
name instead of theirs (cf. Broun, 1994a, 22-3). 

I do not propose to choose between these solutions to the puzzle. Part of its 
beauty is that they could both be right up to a point. To introduce two 
considerations brought into the debate by Katherine Forsyth' s recent work. Her 
discovery that the inscription on the Dupplin Cross commemorates 'Custantin 
filius Fircus '(1995) would seem to favour a ' Scottish' persona, though only if 
it is still seen as a fundamentally post-Pictish monument, and if of course it was 
put up by him, not in his honour and memory. 6 On the other hand, her 
forthcoming paper on the 'Pictish' royal names in the Lindisfarne Liber Vitae 
shows that there is still a hint of Pictish orthography in its spelling of Uoenan' s 
name, if not of Constantine 's own; this therefore backs up the lesson of the 
survival of Pictish spelling in the Icing-lists down to the mid-century, the point 
made by Dauvit Broun (1994a, 22), and one further reinforced by the fact that 
Constantine seems to have given his son the decidedly Pictish name of Drest. 

The point I wish to make is that, instead of trying to adjust the sources into 
a pattern that makes more sense of them, we should ask why they seem to be 

•S» Britain 

talking nonsense in the first place. Is it not on the cards that, when we deduce 
that the scene was set for Kenneth mac Alpin by his immediate predecessors 
one way or the other, this was the very impression that the evidence is trying 
to give us? Whatever the circumstances in which Kenneth's cousins (if they 
were his cousins) took power in Fortriu, it was certainly in his interest to stress 
that Pict and Scot could be represented as having had much the same rulers for 
the previous half-century. There are several reasons why surface appearances 
mightprompta 'double-take'. For one thing, the 'X'/'Y' lists have clearly been 
tampered with, so as to secure the royal title of Kenneth's father, and such 
lightning could strike more than once. The disqualification of this record leaves 
only the 'Synchronisms' and Duan as evidence that Constantine, Angus and 
Eoganan ever were kings of Scots as well as Picts; it is not certain that they 
represent two witnesses rather than one; and even if they do, information 
relayed to Ireland by an eleventh century Scottish establishment should not be 
regarded as foolproof. 7 In any case, the several signs of garbling in these 
sources' sequences just before Kenneth (Table, n. (iv)) are consistent with an 
undercooked tradition. 8 

In addition, much of the remaining evidence is at pains to uphold a 
'continuity' thesis. Even the most 'Pictish' king-lists ('P') reach us only as 
continued by lists of their Scottish successors. At the eleventh century stage 
conveyedbythe 'B' andand'C lists, there is no hint ofa break at Kenneth (col. 
4); and though the Synchronisms make the remark quoted there, nothing 
portentous is said by the solidly Gaelic Duan Albanach either (col. 4). As Mrs 
Anderson has pointed out (1 973, 78-9), it looks as though the 'B '/' C list once 
terminated not with 'et Bred' but with 'et Custantin fil. Cinasda': in other 
words, it ran on uninterrupted for the first three Scottish kings, crossing the 
critical gulf of a new generation. And although list 'A' cuts off in favour of a 
chronicle with a rather different burden (col. 2), as we shall see, several scholars 
have noted (most recently, Broun, 1 994b, 40-5) that 'Pictish' terminology stays 
consistently in place all down col. 2 until 900, before at last giving way to 
'Albania' . Just to show that this is no chance, exactly the same feature is found 
in col. 1 , the strictly contemporary Annals of Ulster, with whose compilers, as 
Hudson persuasively suggests (1994, 55), it is a fair guess that Kenneth's 
spokesmen had their contacts. 

All this is to say, then, that we cannot decide for sure who truly engineered 
the Scottish takeover of Pictland. Equally, we are scarcely entitled to rej ect the 


Da %ae 
l l e Britain, 

dominant version relayed by our sources merely because other texts tell a story 
that may be every bit as artificial. On the whole, it remains somewhat easier 
to believe that Kenneth played upon the precedent of residually ' Scottish' royal 
Picts than that Scottish forerunners were slcilfully reprogrammed as pseudo- 
Picts. 9 In any event, it was the careful spatchcocking of the two traditions that 
has made the whole issue impenetrable. Continuity was at a premium. In the 
early medieval West it very often was. 

Here, my parallels from elsewhere in Britain and Europe first come in. The 
most glaring case of the use and abuse of continuity is the rise, a century and 
a half later, of Brian Boruma (Byrne, 1973, 1 1). He began as ruler of the Dal 
Cais, a petty dynasty from County Clare (like another family of moment in Irish 
history, the de Valeras). He ended as imperator Scottorum, as his secretary 
wrote in the venerable Book of Armagh. Along the way he was genealogically 
linked to the Eoganachta, Munster' s traditional rulers . The link is patently false 
- patently, because of the unrivalled richness of the Irish genealogical record. 
But it conveyed a political truth, which is what mattered. Again, at much the 
same time as Scots were gaining power in Pictland, the dynasty of Merfyn 
began its spectacular ascent by seizing the throne of Gwynedd (Davies, 1 982a, 
105-7). Rhodri Mawr, his son, took over Ceredigion and Powys, while Hywel 
Dda, his great-grandson, swallowed Dyfed. Each move was legitimized by 
inter-marriage and descent in the female line. That apart, the scenario was lent 
a certain familiarity by putting about a story of how another warlord had come 
to power in Gwynedd four centuries before, naming its sub-kingdoms after the 
sons to which he assigned them. The story of Cunedda was good enough to fool 
all historians before David Dumville (1977b, 181-3). On more familiar West 
Saxon ground, Ecgberht, King Alfred's grandfather, claimed rule of both 
Wessex and Kent by family right (Scharer, 1996). He may have been justified 
in either respect; hardly however in both. Last come the Carolingians 
themselves, arch-exponents of the 'consensus putsch '. Merovingian blood had 
turned up in their veins by the ninth century; well before that, Charlemagne 
gave the names of his greatest predecessors to two legitimate sons and one 
bastard (W allace-Hadrill, 1 97 1 , 1 06). It is hard to say who was really taken in 
by all this. It may be wrong to ask. The rhetoric of familiarity met a deeply felt 
demand. Elites never far from the threshold of brutal violence needed to sense 
that their politics had its formal aspect. Sources lie, those for ninth century 
Scotland most probably included; they lie because in a way they were expected 
to. 10 


And so we come to my second theme: violence. In this instance, it may help 
to begin with the wider scene, a short discourse on the use of force in the early 
medieval West. I suspect that, as English-speakers, we are at a disadvantage 
here from the domination of our consciousness by the most notorious of all 
medieval conquests. Yet Hastings was one of extremely few politically 
decisive battles between Adrianople and Bouvines. When Charlemagne 
marched on Lombardy in 774, opposition expired with barely token resistance. 
Lombard institutions were in general respected; but one has to search quite hard 
among the ranks of those running them under Frankish rule to find a scion of 
the old Lombard aristocracy (Wickham, 1981,48,73-4). When the Avars were 
wiped off the map by a brilliantly coordinated campaign, violence was 
apparently limited to the overrunning of their great 'Ring' or stockade; and it 
was the distribution throughout Europe of the hoard of treasure they had 
accumulated over centuries that signalled the evaporation of their power (Pohl, 
1988, 3 12-23). In an age when armies (or the effective parts of them) fought 
for plunder not pay, the enemy' s treasury was the prime military target (Reuter, 
1985, 1990); the richer it was, the farther it went towards erecting a substitute 

Moving to Britain, West Saxonkings fought no battles between 910 and 937 
when establishing an ascendancy over what became England; it is a point of 
some relevance to us that much of the course of these blistering campaigns 
would be wholly unknown, but for the survival of a lone text of the Anglo-Saxon 
Chronicle (Wbitelock, 1979, 213-18). One piece of evidence, however - just 
one and surviving once more by the merest chance - reveals that those who did 
not come instantly to terms with the march of Edward the Elder' s army forfeited 
their lands ipsofacto (Blake, 1962,98-9). Ifwe return to the Gwynedd dynasty, 
the Scots' closest contemporary counterparts, we find fewbattles marking their 
progress in either the Welsh Annals or Asser (Morris, 1980, 48-9, 89-91; 
Keynes and Lapidge, 1983, 94-6): feeble authorities maybe, but a significant 
improvement upon anything from Scotland. The last thing that we are therefore 
entitled to conclude from the failure of ninth century Scottish sources to give 
plausible accounts of military encounters is that Scottish control was basically 
achieved peacefully. It may not even be wrong to rule out of all consideration 
the later legends - alreadyknowntoBerchan,col.3 - of dirty tricks at the Scone 
dinner table. Peter Heather reminded this assembly two years ago that abuse 
of hospitality was a standard tactic in Roman or tenth century German handling 
of obtrusive neighbours (1994, 60). 


What Kenneth mac Alpin (or Constantine mac Fergus) would have had to 
do to establish their supremacy, if Charlemagne's pulverization of the Saxons 
is anything to go by, was to sustain an unusually persistent version of what the 
Irish Annals almost annually characterize as a 'hosting'. Keeping up the 
pressure across Scotland's central belt and beyond, targeting fort after fort, 
rewarding submission and disposing ruthlessly of the recalcitrant, would in the 
end demoralize Pictish opposition as it did others exposed to similar strategies. 
But do we have grounds to think the Scots capable of such campaigning? Two 
years ago, Dauvit Broun thought that possibly 'there was not much left of Dal 
Riata after a couple of decades of Norse settlement in the Western Isles and 
Western sea-board' (1994a, 22). I think it possible that the truth may in a sense 
bethe opposite of this. Ihastento say thatlacceptDr Hudson's elegant removal 
from the stage of nearly all props that have lent verisimilitude to the image of 
Viking rulers as collaborators with the Scots (1994, 40-2, 46, 89, etc.). What 
gives me pause about his going on to argue that they were the enemy of Scottish 
expansion and vigorously combatted, as constantly stressed by our nearest to 
a reliable record, the 'Poppleton' Chronicle (col. 2), is once again comparison 
with Wales and Ireland. 

With Brian Boruma, we are as usual on well-charted ground. Until quite 
recent times, Irish histories still celebrated Brian as the hero of a national 
crusade against the Vikings. We now know that the reason this seemed so true 
for so long was that his dynasty put out a telling piece of propaganda to that 
effect: the 'War of the Gaedhil and the Gaill (Irish and Foreigners)' (Todd, 
1 867). As a matter of fact, Brian owed much of his success to an ability to milk 
the fleets, levies and revenues of Viking towns under his sway (6'Corrain, 
1 972, 120-3 1). Further to that, Professor Wendy Davies has adduced the high 
probability ofViking participation in the vicissitudes of Gwynedd politics from 
the second half of the tenth century (Davies, 1990, 56-60). Thomas Charles- 
Edwards now draws my attention to (late) evidence that the dynasty of Merfyn 
itself came from the Isle of Man (Charles-Edwards, 1995, 706-7). If this is 
right, it may have a real bearing on how we think about the shape of things 
furthernorthinthewestBritisharchipelago: in 'Sodor (Dal Riata)' (his phrase) 
as opposed to Man. From the first appearance of Norse freebooters in the 'Irish 
Sea province', a reservoir of military energy began to build up, its power at the 
disposal of any warlord who found promising channels for its outflow. The 
triumph of the Scots might then be a first manifestation in British history of the 
formidable punch packed until early modern times by the Lordship of the Isles. 

»ff Britain 

Judging by the Irish parallel, the Poppleton Chronicle's impression of an epic 
struggle with Vikings should not be allowed to exclude the possibility. In this 
connection, two final observations can be made. First, Scots knew from the 
start what King Alfred had to learn from experience: the value when dealing 
with Vikings of ships. In the seventh century, the muster of the fer nAlban was 
already in part naval (Bannerman, 1974, 43, 49). Second, the overall effect of 
Vikings in the British sphere and beyond was up to a point to recreate the hectic 
conditions of the immediately post-Roman 'Heroic Age' . As well as furnishing 
recruits for warbands, they unleashed new supplies of the treasure that fused 
warbands together, by liquefying so much of the precious metal that had been 
cast into church plate over the previous four centuries. The Scots may have 
been among the beneficiaries of a markedly destabilized ninth century scene - 
whether or not they were among the few who rode the Viking tiger without 
ending up inside it. 11 

My third theme does concern more peaceful aspects of the exercise of early 
medieval hegemony; or rather, an ideology which could be as merciless in its 
application and as devastating in its effects as any hosting. We can begin with 
legislation. It hardly needs saying today that the making of law (whatever this 
actually amounted to) was seen by the Carolingians as an integral part of their 
via regia. Just as well known is that Alfred's dynasty signalled its new political 
and governmental consciousness by a more or less sustained legislative 
programme, launched by a lawbook of Alfred's own which was essentially a 
restatement of his people ' s law within a new ideological framework (W ormald, 
1977, 128-32). More directly relevant in understanding the Mac Alphas is that 
there is reason to think that Celtic kings were affected by this trend, though their 
efforts survive barely if at all. So Brian Boruma is said by the Annals of 
Inisfallen to have taken hostages 'as a guarantee of the banishment of robbers 
and the lawless' (Mac Airt, 1951, 167). The Annals show his successors 
making laws (cdna) against theft (6 'Corrain, 1974,22-4). Only the Annals let 
us know this, because that sort of law was not kept among the voluminous 
memorials of early Irish jurisprudence. The situation in Wales is yet more 
intriguing. Welsh lawbooks from the high Middle Ages all stand in the name 
of Hywel Dda, the most powerful king of the tenth century. This is not merely 
a matter of a predictably bogus prologue. Hywel' s intervention is signalled 
often enough throughout the texts themselves to leave little doubt that he or his 
authority had once played a major role in the process (Jenkins, 1986, 1,52,94, 
110, etc.). Yet Welsh law comes out like Irish, as lawyers' tracts. Not one 


enunciation canbe confidently ascribed to Hywel. In this light, it seems entirely 
credible that, as Stuart Airlie stressed last time around (1994, 34-6), Donald I 
marked his dynasty's new eminence by a great legislative jamboree (col. 2), 
reissuing the law of a Dal Riatan predecessor; it is certainly not to be 
disbelieved in overreaction to the musings of Fordun or even later 'authorities' 
(HaddanandStubbs, 1878, 122-4(!); Duncan, 1993; Anderson, 1982 122-3- 
Wormald, 1986, 168-9). 

One sort of characteristically Celtic lawgiving may be especially important. 
Here, it is best to begin by standing back, for one last time, in search of the 
widest possible vision. In early medieval political discourse, 'reform' was the 
other side of the 'continuity' coin. It was naturally the one that was uppermost 
more often as the passage of time made it less necessary to stress the lack of 
change. But that does not mean that it was not focal for the self-image of new 
regimes from their inception. It is of course a much more palpable Carolingian 
theme than continuity (Ullmann, 1969; McKitterick, 1977). Carolingians 
never forgot - were never allowed to forget - that they owed their crown to 
the sanction of the Pope and of a Church who looked to them for Josiah-like 
redress of the unregeneracy that had so angered God as to unleash the forces of 
Islam. With the West Saxons, it all began at a humbler level. But the 83 8/9 deal 
which set up the succession (and most probably unction) of Ecgberht's son 
^thelwulf, thefhstfather-sonsuccessionin Wessexfortwo centuries, involved 
concessions to Canterbury and probably Winchester (Brooks, 1984, 197-203). 
By the eleventh century, Archbishop Wulfstan was outpacing the Carolingians 
themselves in his zeal for an ordered and holy society. It would be hard to find 
such active concern in Celtic kings. Still, Irish rulers were induced to grant 
exemptions to churches general and specific from the various burdens of 
secular government; so Aed 'the Ordained' was persuaded in 804 by a synod 
under the Abbot of Armagh to free churches from supplying recruits for his 
'hosting'; and Brian in 1012 freed 'all St Patrick's churches' from secular 
imposts (Byrne, 1973, 159-60; 6'Corrain, 1972, 127; 6'Riain, 1990). That 
policy at least was of a piece with Europe-wide awareness of the value to kings 
of cultivating the goodwill of saints, above all those who were in any sense 
'national' saints. Carolingians adopted the 'special patronage' of St Denis, 
^thelwulf lavishly re-adorned the shrine of Aldhelm, the West Saxon royal 
saint and scholar (Hamilton, 1870, 389-90); his later successors earnestly 
solicited the supportofStCuthberhtfortheirnorthern strategy (Rollason, 1989, 
144-52). Brian was rewarded for his services to St Patrick with the unparalleled 


honour (for a non-Ui Neill layman) of burial at Armagh (Byrne, 1 973, 256-7). 
Even in Wales, the poet/prophet of Armes Prydein looked to the intercession 
of 'Dewi and the saints of Britain' to put 'the foreigners' (English) to flight 
(Williams, 1982, 4-5, 8-9). Carolingian and sub-Carolingian kings all had a 
Kirchenpolitik. It was then duty and their interest. 

The relevance of all this for the earliest Scottish Icings emerges from 
excerpts in the Table. It is part of a distinct if subtle shift of emphasis as between 
the twelfth century ('X'/'Q') records in col. 5 and the eleventh century evidence 
of col. 4 that more is made of a new deal for the Church (though not without 
finding suitable precedents, as when the foundation of the nodal shrines of 
Dunkeld and St Andrews is credited to the key figures of Constantine and 
Angus, sons of Fergus). Thus, Giric 'first gave freedom to the Scottish Church 
which was under servitude until that time after the custom and manner of the 
Picts' . It is reasonable to connect this with the freedoms granted by Irish kings 
(Cowan, 1981, 1 1), but with the critical riderthat a new dispensation is this time 
linked to a drastic change of secular regime. The standing of the twelfth 
century records is unfortunately far from secure. But Giric' s contribution is 
startlingly borne out by the mysterious 'Dunkeld litany' (Table, ad fin.), with 
its very Carolingian invocation for 'his army' . The 'reform' theme also comes 
out clearly in a much better source, the Poppleton Chronicle (col. 2). Though 
even this can be no earlier as it stands than the 970s, it may, to repeat, be wrong 
to regard it as purely retrospective justification for change that was in reality 
muchsmoother. Thethemes of 'Reform' and 'Continuity' were complementary 
in images of Carolingian and post- 1066 change; and each is in different ways 
stressed by the Poppleton Chronicle itself. Thus it may well say something 
significant about the spirit in which the Scottish takeover was conceived from 
its very outset that it includes a counterpart to Giric 's concession, though 
ascribed to Constantine II and his bishop. Even more suggestive is a passage 
whose seminal importance was spotted by Ted Cowan (1981, 14): 'God 
deemed (Picts) deserving of being deprived of then inheritance by reason of 
their wickedness, because they not only spurned the mass and commandment 
of the Lord, but in right of justice would not be put on a level with others'. It 
does not matter what (if anything) the Picts had actually done wrong. Here is 
the hiss of the most lethal weapon in the early medieval ideologue ' s arsenal: the 
image of a people expelled for its sins from its promised land. 

I have recently argued that B ede and Alfred provided the ideological charter 


of a new English kingdom by adapting the Israelite model to Anglo-Saxon 
experience of the Britons and the Vikings (1994). And yes, I now venture the 
same proposal for the Scots, their compeers in ninth century statecraft. The 
history of the Scots diverged from that of then fellow-Celts at this point, with 
all that that was to mean for the future of these islands, because they did what 
the Welsh or Irish never quite managed to do: they harnessed a compelling 
political idea. They could represent themselves as instruments of God' s wrath 
at the sort of backslidings which, in Gildas view as rehearsed by Bede, had once 
cost Britons their homeland and would now exact the same cruel price from 
Picts. But the Scottish picture has an intriguing twist. The role provided for a 
patron saint was in this case so central as to make Denis, Patrick or David seem 
peripheral. Archie Duncan proposed at length and with force (1981) that, 
whether or not Columba was the Apostle of the Picts, the Picts themselves 
believed as much by the time they got in touch with Monkwearmouth-Jarrow 
over the Easter issue. But Columba was a Scot, not a Pict; he was above all a 
patron of Dal Riata. What was more, the Picts had ignominiously sent his 
clergy packing back in 717. So Scots could also portray themselves as the 
agents of his irate return. And that is just what they did. One of the Poppleton 
Chronicle's rare snippets on Kenneth (col. 2) is his translation of Columba's 
relics to a church he built, probably Dunkeld. Stuart Airlie has eloquently 
expounded the logic of such moves (1994, 36-41). Alan MacQuarrie (1992, 
122-3) notes the implications of a Dunkeld cross-slab, replete with martial 
imagery ('exercitu suo'), and with echoes of Columban reformist sculpture in 
Ireland. Meanwhile, a non-ecclesiastical perspective is given by col. 3 . Amidst 
allBerchan's gobbledygook, one message comes over loud and clear. Kenneth 
and his successors are treading where Columba (and Aedan mac Gabrain) had 
trodden before. A saint comes unto his own. This time, they have no choice 
but to receive him. 12 

Thomas Clancy's discoveries about Abbot Diannait of Iona con fir m my 
long-held suspicion that more might be unearthed along these lines by study of 
the mind-set of the Cell De reform. If the persistence of the word means 
anything, it may have had yet more influence in ninth century Scotland than in 
Ireland. 13 There, according to Francis- John Byrne (1 973, 1 57-8), it propagated 
ideals of Carolingian kingship. Dr Clancy's argument speaks for itself. I shall 
conclude by returning more or less to where we started. By the second half of 
the ninth century, the idiom of Irish churchmanship was substantially, even 
aggressively, vernacular; this may be especially true of the Celi De (e.g. 

Dark ^ 

Kenney, 1929, 468-82). Now, when rightly insisting at the 1993 symposium 
that analysis of ninth century Scotland had to reckon with the disappearance of 
Pictish, Dauvit Broun proposed the disruption of traditional social structures 
caused by the Vikings as a possible factor (1994a, 27-30). The difficulty here 
is that it is not obvious why the Vikings (or even any 'social change' arising) 
should have been so much more subversive of Pictish Ps than of Gaelic Qs. 
However, if the Scottish conquest were bound up with the strident assertiveness 
of a reforming and emphatically Gaeliphone Church, then it is possible to see 
how Pictish couldbe marginalized. A new ethnicity would coalesce around the 
tongue in which Columba entertained his string of angelic guests. It may not 
be chance that Constantine II was the hero, after Kenneth, of Berchan as well 
as the chronicles: just in the time of this would-be Celi De abbot, we begin to 
hear of Alba. Against that setting, the displacement of Gaelic culture from the 
Scottish establishment after the twelfth century would create the semi-prehistoric 
mists though which we must all peer. It is one good answer to Kathleen 
Hughes's question, 'Where are the writings of early Scotland?' (1980), that 
many were lost when Gaelic, like Columba, went out of fashion. And by the 
time that Scots were again preoccupied with a literate vernacular, it was not, 
very much not, the one they had spoken half a millenium before; there was thus 
no place for a Scottish Archbishop Matthew Parker. 

Over the last two decades, I have been learning that extrapolating from 
England's history is not necessarily the best way to understand Scotland' s. For 
all that, I am prepared to say that aspects of '1066 And All That' may not take 
us too far from what happened in 842. Like the Normans or the West Saxons 
and Carolingians before, Scots could claim a legitimate title to Pictland from 
recent political history. Like them, they could seem to create a climate under 
which the Church might live in a new freedom; and if we can envisage the 
Carolingian or Lanfrancian reforms powered by a far more self-confident 
vernacular than Alfred' s or Wulfstan' s English ever was, we may get some way 
towards realizing their impact. In any event, just because we know so little 
about him is no excuse for denying Kenneth mac Alpin the role of conqueror 
assigned to better documented 'Dark Age' hegemonists. It is what the record 
does ultimately invite us to do. I am ready to say, then, that the Picts were after 
all conquered by the Scots in the ninth century. Their aristocracy suffered in 
the same way as others, the English of 1 066 not least, from the advent of rulers 
prepared to stress all forms of continuity except the locus of the incumbent elite 
(place-names with 'Pit-' prefixes to Gaelic personal names always seem to me 

■ Scotland 
«» BrUatr, 

to speak volumes); rulers who simultaneously or soon came to see their role 
as the scourge of God on a corrupt ecclesiastical establishment in the name of 
the saint who first brought the Faith to North Britain. Guesswork? Well, yes. 
But guesswork which predicates nothing for Scotland' s history not attested for 
similar societies under comparable conditions. And guesswork which, whatever 
its other faults, does have the merit of imagining that the first Celtic culture to 
die out in these islands did not do so without a fight. 

■ Christ Church 
University of Oxford 


Tke Emergence of tke Regnum Scottorum 

1 . This foray into what has been wisely described as 'one of those refreshing 
issues about which complete agreement is impossible' (Broun, 1994a, 21), is 
inspired by the experience of fifteen years' teaching at the University of 
Glasgow and attendance at the 'Scottish Medievalists'; in particular, to what 
I learned from working with Professors Archie Duncan and Leslie Alcock; and, 
more crucially if more obviously, to a life shared with a third distinguished 
historian of Scotland. Apart from those general debts, I am most grateful to 
Katherine Forsyth for generously sharing her already almost unrivalled 
knowledge of Pictish problems, and for further advice to Dauvit Broun, 
Thomas Charles-Edwards, Roger Collins, Rees Davies and Archie Duncan. 
However, it follows from Dr Broun' s caveat that none of these scholars is 
answerable for my paper's lingering perversities. I say little of archaeology, 
art-history or place-names: not only are they covered by other essays in this 
series, but I am even less equipped to dabble in such depths than in the eddies 
of the written sources; the parameters of artefactual and linguistic research 
have anyway tended to be set by understanding of more 'purely' historical 
evidence. In the text, though not when it matters in the table, I have usually 
anglicized the spelling of Celtic names, so as to retain an element of recognition 
in what is already regrettably rebarbative. 

2. The pioneer of this approach to the problem seems in fact to have been 
Watson, 1926,218-34. 


'« Britain, 

3 . 1 would add that if Dr Hudson' s book seems to attract more than its fair share 
of my critical attention, that is a tribute to his immensely stimulating treatment 
of the subject. 

4. Too much should not, however, be made of the 'Fortriu' label in the early 
ninth century: it was used for such unimpeachably (and exclusively?) Pictish 
kingships as that of Bridei s. Bile (Mac Airt and Mac Niocaill, 1983, 154-5); 
and cf. ibid., 216-17. 

5. It will be evident that I accept the case for Pictish matriliny (Miller, 1982; 
Sellar, 1985), at least in so far as the change that seems to come over the Pictish 
Icing-lists in the ninth century is best explained in this way. However, the 
exploitation of Lex Salica in the fourteenth century by French lawyers bent on 
excluding the succession of Edward III serves as a reminder that such ' customs ' 
can just as well be rooted in recent political pressures as in immemorial 

6 . Ms Forsyth tells me that she herself regards this as ' a Pictish monument from 
the opening decades of the ninth century'; this is what is indicated by the art- 
historical evidence, the use of Latin rather than Irish, and 'the very significant 
Anglo-Saxon artistic influence'. The general impression is that it is both 
'eclectic' and 'up-to-date', 'an import from Gaeldom in the latest style' ; 'The 
representation of a Pictish name (Fircus) in Gaelic orthography would fit in 
with this'. For another recent assessment, see Alcock (1992), 238-41. 

7. As noted by Boyle (1971, 170), there are signs that the 'Synchronisms' were 
in some sense written with Scottish sensitivities in mind. By the same token, 
no other Irish source can easily be taken as confirmation of the ' truth' about the 
Dal Riatan ascendancy: Hudson, 1994, 30-1, 54-5; cf. his point cited below 
about Cenel nGabrain influence on A U, transmitted via the contacts that each 
had with the Ui Neill, among the implications of which is that the Irish Annals 
may conceal the element of violent disruption in the Scottish takeover. 

8 . This is not to say that Kenneth' s own descent is fraudulent: he couldperfectly 
well have been ' son of Alpin, son of Eochaid, son of Aed Finn, etc. ' , as claimed 
by the apparently tenth-century ' Genealogy of Alba' (Bannennan, 1 974, 65-7), 
without Alpin ever having been an acknowledged Icing. 


9. Perhaps the best case for Constantine's 'Scottishness' is a name which, if 
Alan Macquarrie is right (1990, 10-13), is likelier to have been given to a man 
schooled for Scottish rather than Pictish rule. 

10. It is a drawback of Mrs Anderson's ground-breaking, scholarly, and often 
inherently plausible argumentation (1973, 1982, etc.) that it was conceived 
before the 'genealogicalrevolution' ofthe 1970s (best epitomizedbyDumville, 
1977a) made its effects felt. 

11. This interpretation of conditions on the western seaboard has an obvious 
bearing onDr Hudson's suggestion (1994, 56-7, 128-47) that the Cenel Loairn 
penetrated the Great Glen at much the same time as the Cenel nGabrain made 
their way into Strathmore, providing kings Giric and Constantine in the ninth 
century, and Macbeth and Lulach in the eleventh. The Macbeth and Lulach 
genealogies (O'Brien, 1962, 329-30) may very well not be authentic, while that 
of Giric is as late as Fordun. But contemporary assaults by freebooting bands 
on the north-east seem entirely likely, whether or not under Cenel Loairn 

12. For the use of Columba's relics in battle, see Airlie (1994), 37-41; Hudson 
(1994), 70; and cf. Smyth (1984), 213-14. Whether or not it is true that the 
dynasty were usually buried on Iona (Cowan, 1981, 7), it is a fact that four of 
its members were named 'servant of the Dove/Columba'; and Mael Coluim 
was not a common name among the legion of Irish kings. As late as the first 
quarter ofthe twelfth century, Alexander I apparently commissioned a copy of 
Adomnan from an 'insula pontificum' (probably Celi De Loch Leven): 
Anderson, 1961, 10. 

13. One result of Dr Clancy's research is to remove some of the reservations 
about use of the term Celi De in twelfth century contexts that were powerfully 
urged by the late Professor Ian Cowan (1974). 


Tabulated Evidence ~ 

Tke Emergence of the Regruxm, Scottorum 

I Annals of Ulster II Poppleton Chronicle IIIBerchan 

W King-lists 

V King-lists X/Q 

781 Death of Fergus 
m. Ecach (= Eochaid) 

782 Death of Dub 
Tholarg 'rex 
Pictorum citra 

Pictos', Conall m. 
Tadc defeated, 
Constantine victor 

792 Death of 'Donn 



807 Killing of 
Conall m. Tadc by 
Conall m. Aedacan 

820 Death of 
'Custantin m. 
Fergusa, rex 

834 Death of 
'Oengus m. Fergusa, 

839 'Genntib' 
victory over 'men 
of Fortriu', death 
Oengus, Bran m. 
Oengus, Aed m. 
Boanta& 'almost 

[Talorgen f. Onuist] 

[Castantinf. Wrguist] 

([Columba] will be 
a scholar, a seer, a 
sage of the son of 
God ... a warrior and 
cleric, pure and 

fierce He will 

not be absent in 
Iona (though he 

[Aedan m. Gabrain] 
will cast the Picts 
into insignificance 
He is the first man 
who will rise in 
the East ) 

[Unuistf. Wrguist] 

[Drest f. Constant™ 
et Talorgen f. Wthoil] 

[Uuen f. Unuist] 

(Fergus </> Eoclwidh) Fergus f. Hethfin 
[Alpin f. Heoghed] 

Talorcen f. Oinuist 


858 Death of 

Kenneth m. Alpin Scottorum rexit 
'rex Pictorum' 

Son of the clan of 
[Aedan] will take 
Pictaviam ... quos the kingdom of 

Domnall (m. Cusantin) 

Conall X 2 

Custaintin (m. Fergusa) 

Constantin f. Fergusa 
... 'Hie aedificavit 

Uidnuist f. Uurguist 

Drest f. Constantin 
Aedh (m. Boanta) 
Eoghanan (m. Oengusa) 
Unen f. Unuist 
Uurad f. Bargoit 

? Ailpin (m. Echach) 

Cinaeth f. Alpin 
(Synchr.: he was the 
first king of the 

... 'Hie aedificavit 

Eoganan f. Hungus 
Brude f. Ferat 
Kinat f. Ferat 
Brude f. Fotel 
Drust f. Ferat ... 
'Hie occisus est 
secundum alios 
apud Sconam' 

1*8 Brtt&u 


(i.e. Picts) delevit. 
Deus enim eos pro 
inerito sue malicie 
alienos ... hereditate 
dignatus est facere: 
qui illi non solum 
Domini missam ac 
preceptum spreverunt, 

isequi ' 
parari <n>oluerunt ... 
Antequam veniret 
Pictaviam Dalriete 
regnum suscepit. 
Septimo anno regni 
sui reliquias Sancti 
Columbe transportavit 
ad ecclesiarn quam 



Scotland ('righe 
Alban'). He is 
the first king that 
will reign in die 
East from among the 
Erin in Alba ... after 
violent slaughter. 
The fierce men in 
the East are deceived 
by him ...a deadly 
pit(?) death by 
wounding in the 
middle of Scone of 

Gael who possessed 
the kingdom of Scone) 

...Long will it be 
till his like will 

862 Death of Domnall In huius tempore jura 
m. Alpin, 'rex ac legis regni Edi f. 

Pictorum' Ecdach fecerunt 

Domnall f. Alpin 

Kinafh m. Alpin ... 
Super Scotos 
regnavit, destructis 
Pictis ... sepultus 

in terrain Pictorum 

Goideli cum rege suo ... 

875 Battle of 

... vastavit Amlaib 

Another young king 

et Custantin f. 

... interfectus est a 

'Picts' vs. 'dark 

cum gentibus suis 

will take sovereignty 


Norwigensibus ... 

foreigners', with 


the cow-herd of die 

sepultus in Iona 

'great slaughter' 

bello ... in Dolair 

byre of the cows of 

inter Danarios et 

the Cruifhnech ... 

876 Death of 

Scottos ... Normanni 

three battles will 

Custantin m. 

annum integrum 

be gained over the 

Kenneth 'rex 

degemnt in Pictavia. 

heathen ... (five) 


years as 'Ri Alban' 

878 Aed m. Kenneth 

Edus ... 


... interfectus ... a 

'rex Pictorum' 

Girg f. Dungal, 

killed 'a sociis' 

sep. in Iona 

Eochodius ... 

Ciricius ... 

[positive view of 

Giric m.Dungaile 

Hie primus dedit 

Giric, scathing on 


900 Death of Domnall 

Donald f. Constantini 

Donald twice over] 

Domnull f. Constantin 

ecclesite Scoticanai 

m. Constantine, 

...Normanni turn 

qua: sub scrvitute 

'ri Alban'. 

vastavemnt Pictaviam 

erat usque ad 

bellum ... inter 

ilium tempus ex 

Danarios et Scottos 

constitutione et 

more Pictorum 


Welcome, welcome if 

Custantin f.^da 

■ Hie dimisso regno 

Edii ... Normanni 

he it is who has 

... abbas factus 

predaverunt Duncalden 

long been prophesied 

Keledeorum S. 

omnemque Albaniam .. 

... A fair long reign 

Andres ... et ibi 

...Ac in vi anno 

with fruit, ale, music 

... sepultus 

Constantinus rex et 

... Battles will not 

1 * 9 Britdm 

Cellachus episcopus stand against him ... 

leges disciplinasque God is faithful 

fidei atque iura to him ... 



pariter cum Scottis 

in colle credulitatis, 
952 Death of prope regali civitati 

Constantinem. Aed, Scoan devoverant 
'ri Alban' custodire 

From Tke 'Dunkeld Litany 

Ut animalia nostra ab omni lue pestifera custodias Ut Episcopos, 

Abbates Kiledeos et omnem populum totius Albaniae, conserves et protegas. 
Ut regem nostrum Girich cum exercitu suo ab omnibus inimicorum insidiis 
tuearis et defendas: TE ROGAMUS AUDI NOS. 

On, Tabulated Sources 

For reasons given at the outset of the text, it seems important not only to set out 
as much as possible of the exiguous evidence for the course of events in ninth 
century northern Britain, but also to bring out the extent to which different 
sources may have different agenda: hence the use of five columns, except for 
the entry from the 'Dunkeld Litany', quoted in conclusion. 

(i) Col. 1: Annals of Ulster (AU: Mac Airt and Mac Niocaill, 1983); I have 
not cited other witnesses to the 'Chronicle of Ireland' (Hughes, 1972, 99-1 15; 
Smyth, 1972; mostrecently, GrabowskiandDumville, 1984, 1 1 1-27), inasmuch 
as they add little or nothing to what A U has to offer on North British matters. 
Everything we know about^ [/suggests that it may be taken as a contemporary 
(which is not of course to say objective) record; but this applies only to entries 
in the main hand, not to glosses or interpolations. 

(if) Col. 2: 'Poppleton Chronicle' (ed. Anderson, 1973, 235-60; see also 
Anderson, 1949,31-42). This Chronicle stands in the MS as a continuation of 
Pictish king-list 'A', one of the 'P' class (cf. (iv) below), with a red-ink initial 

Scotland , 
ȣP Britain 

that marks no more of a break than those denoted by newparagraphs throughout 
Mrs Anderson's edition, 245-53 - and with less of a break than at the 
introduction of the Scottish king-list proper (253, n. 161), which is at least 
launched by a red-ink rubric: cf. the facsimile in Skene, 1 867, facing p. 3 , also 
Anderson, 1949, 34-7; accordingly, this column also contains in square 
brackets entries from the 'A' list after the 780s, retaining MS orthography as 
reproduced by Mrs Anderson, but omitting the attributed reign-lengths on the 
grounds that, in principle if not in fact, these are yet more liable to distortion 
than the order of names. For the unique authority of this source see Cowan, 
1981; Miller, 1982, 137-42; andBroun, 1994b; with (forthcoming) a study by 
JohnBannerman, Studies in the History of Scotia, and an edition/translation by 
Ben Hudson. I accept the case that it was compiled (in Gaelic) in the later tenth 
century from materials that were contemporary for the ninth; if this makes it 
relatively reliable in the facts it chooses to report, the same cannot necessarily 
be saidforthe choice itmade, norforthe gloss itputuponthem. Theprima facie 
conclusion from its final entry is that it was written not only in the reign of 
Kenneth II (97 1 -95) but also at Brechin. However, there is an obvious case for 
. an Abernethy element among its sources, given its inclusion of two Abernethy 
'charters' (Davies, 1982b, 273), only the first of which is shared with lists 'B' 
and 'C (col. 4); and it might paradoxically be concluded from the failure to 
specify that Dunkeld was the new ecclesia to which Kenneth Mac Alpin 
translated St Columba' s relics (Macquarrie, 1992,121-3), that Dunkeldprovided 
another ingredient in the mixture. Less likely is any significant St Andrews 
contribution. Part of the Poppleton collection probably did originate there, and 
a whole textual family of the king-list on which it drew has strong St Andrews 
connections (see col. 5). But the Chronicle's perspectives differ in a number 
of ways from that set of lists, e.g. in its lack of enthusiasm about Giric (p. 142 
above), and its reserve verging on hostility towards Constantine II' s side of the 
dynasty (cf. Hudson, 1994, 91-4) - it does not even mention that Constantine 
II's retirement was to St Andrews. 

(hi) Col. 3: 'Prophecy of Berchan' (ed. in part Skene, 1867, 79-105; tr. 
Anderson, 1922, l,passim, and cf. xxxiv-xxxvi; also Anderson, 1930; anew 
edition/translation is forthcoming from Ben Hudson). The evidence of this 
example of Celtic 'back-prophecy' is not to be accepted uncorroborated, even 
when one canbe confident what it means (note the regular discrepancy between 
Skene and Anderson translations, the latter being quoted here). If it was really 
composed as late as the 1 1 60s (compare Anderson, 1 922, xxxv, with Anderson, 


1930, 4), this is unexpected testimony to the persistence of the 'Gaelic' 
ideology it represents (cf. above, pp. 144). 

(iv) Col. 4: this column combines the evidence of the 'Synchronisms' 
(Thurneysen, 1933; Boyle, 1971) and 'DuanAlbanach' (Jackson, 1957) on the 
kings of Dal Riata and from Kenneth mac Alpha onwards, with that of the other 
'F class Pictish king-lists, 'B' and 'C (Anderson, 1973, 261-3; cf. Miller, 
1982 146-8, 159-61). It is important (cf. above, p. 136) that the latter continue 
with no break through Kenneth and down to Malcolm III, and it is the 'B' list 
that is the major item in this column (again in Anderson's orthography); Dal 
Riatan kings are entered in Italics, spellings being Boyle's, from the Edinburgh 
MS of the Synchronisms, and brackets denoting information lacking in one or 
other versions (these include patronymics, among them the generally discredited 
notion that Domnall was Constantine's son). The Duan and the ancestor of the 
'B V'C lists appear to date to the reign of Malcolm III, and it seems certain that 
they share a common source (Anderson, 1973, 48-9). Less certain but not 
impossible is that this was also the source of the Synchronisms {ibid., 44-6); we 
could thus envisage that a package representing one interpretation of Dal 
Riatan, Pictish and Scottish royal succession made its way to Ireland in the later 
eleventh century (cf. ibid., 51); though we should then have to reject Mrs 
Anderson's view that the short version of the Synchronisms, ending with 
Malcolm II (d. 1 034), was the older, as well as their ascription by Skene to Fland 
Mainistrech (d. 1 056). It should also be noted that, unlike the Duan or even the 
'X'/'Y' lists (below, col. 5), shorter and longer Synchronisms each intrude a 
King Alpin between Eoganan mac Oengusa and the start of the Scottish series 
proper (Boyle, 1971, 170; Anderson, 1973, 46, n. 10); but their texts seem 
corrupt in different ways, the former repeating Eoganan and the latter omitting 
Kenneth himself; this may be further evidence of an adjusted story of the Scots' 
takeover (above, pp. 135-136, below, col. 5). 

(v) Col.5: here, Mrs Anderson's 'X' class ofDal Riatan and Scottish king-lists 
(1973, 49-67) is blended with the Pictish lists of her 'Q' class (ibid., 84-102): 
spellings are those of list 'F' (ibid., 271-5). It needs to be appreciated that, as 
is not immediately apparent from her separate discussions, 'Q' lists are 
invariably incorporated into 'X' lists, usually and perhaps originally between 
theDalRiatankingsandKennethMac Alpin; only one 'X' list does not include 
a'Q' list, and that in fact starts with Kenneth. Most'X' lists have a St Andrews 
connection of sorts (Cowan, 1981, 6-7), and the notice in 'Q' lists of the 


'building of St Andrews' by Angus mac Fergus (IT) seems to link up with tire 
St Andrews foundation legend; it is reasonable, therefore, to trace the 'X'/'Q' 
tradition back to that house. As for its date, much depends on the interpretation 
of Mrs Anderson's 'Y' lists (1973, 49-52, 67-76), a group tending to associate 
in one way or another withMelrose. This group lacks the notes on each Scottish 
reign which are the key feature of 'X' lists, and which are partially excerpted 
in this column; if, as Mrs Anderson argues, they were dropped from the source 
of the 'Y' lists' rather than added to that of 'X', then they must pre-date the 
1160s, when the ancestor of one of the 'Y' lists appears to have become 
fossilized; otherwise, they could be dated at any time down to Alexander II' s 
reign, when the source of one set of 'X' lists evidently concluded. A significant 
aspect of both 'X' and 'Y' lists is that they re-arrange the order of Dal Riatan 
kings, so as to make the Alpin who probably died in the 730s into the immediate 
predecessor of Kenneth (mac Alpin): that is, once again to convert the Alpin 
who fathered Kenneth into a king (compare the Synchronisms, above, col. 4). 
The persistent confusion over this Alpin must tend to discredit the otherwise 
circumstantial evidence of the 'Chronicle of Huntingdon' (Skene, 1867, 209; 
Anderson, 1922, 270-1; Anderson, 1973, 194-5; Cowan, 1981, 13-16; 
Hudson, 1991, 15-17), in that Alpin father of Kenneth is there represented as 
a king, whose death in battle against the Picts Kenneth went on so spectacularly 
to avenge; the Huntingdon evidence is therefore omitted from this table. 

(vi) The 'Dunkeld Litany' (for its provenance, see Haddan and Stubbs, 1 873 , 
278-85) seems to be an ahnost entirely unknown quantity, having been ignored 
since the 1870s until Ben Hudson again drew attention to it (1994, 131) - 
though cf. McRoberts (1953), no. 81. Professors Archie Duncan and Donald 
Watt kindly warn me that anything from such a context is to be regarded with 
the utmost suspicion; yet it is difficult to see that the quoted passages, or some 
other parts of this litany, are likely to have been invented after the twelfth 

'£ 3 Britain 

Airlie, S. 1994, 'The View from Maastricht', in Crawford, 1994, 33-46. 

Alcock, L. and E. 1993, 'Reconnaissance excavations on Early Historic 
fortifications and other royal sites in Scotland, 1 974-84; 5 : A, Excavations and 
other fieldwork at Forteviot, Perthshire, 1 98 1 , etc. ' , Proceedings of the Society 
of Antiquaries of Scotland, 122, 215-87. 

Anderson, A. 1922, Early Sources of Scottish History A.D. 500-1286, 2 vols, 

Anderson, A. 1930, 'The Prophecy of Berchan', Zeitschrift fir celtische 
Philologie xviii, 1-56. 

Anderson,M. 1949, 'The Scottish Materials in the Paris Manuscript, Bib. Nat, 
Latin 4126', Scottish Historical Review xxviii, 31-42. 

Anderson, A. and M. 1961, Adomnan 's Life of Columba, Edinburgh. 

Anderson, M. 1973, Kings and Kingship in Early Scotland, Edinburgh. 

Anderson, M. 1982, 'Dalriada and the creation ofthe kingdom of the Scots', in 
Whitelock et al, 1982, 106-32. 

Bannerman, J. 1971, 'The Scots of Dalriada', in G. Menzies (ed.), Who are the 
Scots?, London, 66-79. 

Bannerman, J. 1974, Studies in the History of Dalriada, Edinburgh. 

Bartlett,R. 1993, The Making of Europe. Conquest, Colonization and Cultural 
Change 950-1350, Harmondworth. 

Blake, E. 1962, Liber Eliensis, London, Camden Society, 3rd ser. xcii. 

Boyle, A. 1971, 'The Edinburgh Synchronisms of Irish Kings', Celticaix., 169- 

l& Bcttatn, 

Brooks, N. 1984, The Early History of the Church of Canterbury, Leicester, 
Studies in the Early History of Britain. 

Broun, D. 1994a, 'The Origin of Scottish Identity in its European Context', in 
Crawford, 1994,21-31. 

Broun, D. 1994b, 'The Origin of Scottish Identity', in C. Bj0rn, A. Grant and 
K. Stringer (eds), Nations, Nationalism and Patriotism in the European Past, 
Copenhagen, 35-55. 

Byrne, F-J. 1973, Irish Kings and High-Kings, London. 
Chadwick, H. 1949, Early Scotland, Cambridge. 

Charles-Edwards, T.M. 1995, 'Language and Society among the Insular Celts 
AD 400-1000', in M.J. Green (ed.), The Celtic World, London. 

Cowan, E. 1981, 'The Scottish Chronicle in the Poppleton Manuscript', Innes 
Review xxxii, 3-21. 

Cowan, I. 1974, 'The Post-Columban Church', Proceedings ofthe Church 
History Society of Scotland 8, 245-60. 

Crawford, B. 1994, Scotland in Dark Age Europe, St Andrews, St John's 
House Papers no. 5. 

Davies, W. 1982a, Wales in the Early Middle Ages, Leicester, Studies in the 
Early History of Britain. 

Davies, W. 1982b, 'The Latin charter-tradition in western Britain, Brittany and 
Ireland in the early mediaeval period', in Whitelock et al, 1982, 258-80. 

Davies, W. 1990, Patterns of Power in Early Wales, Oxford. 

Davies, R. 1990, Domination and Conquest. The experience of Ireland, 
Scotland and Wales, 1100-1300, Cambridge. 

Dumville,D. 1977a, 'Kingship, Genealogies and Regnal Lists', in Sawyer and 
Wood, 1977, 72-104. 

« Baffin 

Dumville, D. 1977b, 'Sub-Roman Britain: History and Legend', History 62, 

Duncan, A. 1 975, Scotland: the Making of the Kingdom, Edinburgh History of 
Scotland I. 

Duncan, A. 1 98 1 , 'Bede, Iona and the Picts' , inR.H.C. Davis and J.M. Wallace- 
Hadrill (eds), The Writing of History in the Middle Ages. Essays Presented to 
Richard William Southern, Oxford, 1-42. 

Duncan, A. 1993, 'The Laws of Malcolm MacKenneth', in A. Grant and K. 
Stxmger(eds),MedievalScotland. Crown, Lordship and Community, Edinburgh, 

Forsyth, K. 1 995 , ' The Inscriptions on the Dupplin Cross ' , From the Isles of the 
North: Medieval Art in Ireland and Britain (Proceedings of the Third 
International Conference on Insular Art) Belfast, 237-44. 

Forsyth, K. 1 996, Language in Pictland: the case against non-Indo-European 
Pictish, Van Hamel lecture, Utrecht. 

Frame, R. The Political Development of the British Isles, 1 100-1400, Oxford. 

Grabowski, K. and Dumville, D. 1984, Chronicles and Annals of Mediaeval 
Ireland and Wales, Woodbridge, Studies in Celtic History IV. 

Greene, D. 1 972. Review of K. Jackson, The Gaelic Notes in the Book of Deer, 
Studia Hibernica 12, 167-70. 

Greene, D. 1983, 'Gaelic: syntax, similarities with British syntax', in D. 
Thomson (ed.), Companion to Gaelic Scotland, Oxford, 107-8. 

Haddan, A. and Stubbs, W. 1873, Councils and Ecclesiastical Documents 
relating to Great Britain and Ireland, 3 vols, Oxford (vol. U). 

Hamilton, N. 1870, William of Malmeshury, Gesta Pontificum Anglorum, 
London, Rolls Series 52. 

*» Britain 

Heather,P. 1994, 'State Formation in Europe in the First Millenium A.D.', in 
Crawford, 1994, 47-70. 

Henderson, I. 1967, The Picts, London. 

Hudson, B. 1994, Kings of Celtic Scotland, Westport, Connecticut. 

Hudson, B. 1991, 'The Conquest of the Picts in Early Scottish Literature', 
Scottish Studies xv, 13-25. 

Hughes, K. 1972, Early Christian Ireland: Introduction to the Sources, 

Hughes, K. 1980, 'Where are the writings of early Scotland?', in K. Hughes and 
D. Dumville (eds), Celtic Britain in the Early Middle Ages, Woodbridge, 
Studies in Celtic History II, 1-21. 

Jackson, K. 1955, 'ThePictishLanguage',inF. Wainwright(ed.), TheProblem 
of the Picts, Edinburgh, 129-66. 

Jackson, K. 1957, 'The Duan Albanach', Scottish Historical Review xxxvi, 

Jenldns, D. 1986, HywelDda, The Law, Llandysul, Welsh Classics 2. 

Kenney, J. 1929, The Sources for the Early History of Ireland: Ecclesiastical, 
New York. 

Keynes, S. and Lapidge, M. 1983, Alfred the Great, Harmondsworth. 

Mac Airt, S. 1951, The Annals of Inisfallen, Dublin, Institute for Advanced 

Mac Airt, S. and Mac Niocaill, G. 1983, The Annals of Ulster (to A.D. 1131), 
Dublin, Institute for Advanced Studies. 

Macaulay, D. 1975, Review of K. Jackson, The Gaelic Notes in the Book of 
Deer, Scottish Historical Review liv, 64-7. 

Darlm e 
l S7 Britain 

Ullmann, W. 1969, The Carolingian Renaissance and the Idea of Kingship, 

Wallace-Hadrill, M. 1971, Early Germanic Kingship in England and on the 
Continent, Oxford. 

Watson, W.J. 1926, The History of the Celtic Place-Names of Scotland, 

Whitelock, D. English Historical Documents (vol. I), c. 500-1042, 2nd edn, 

Whitelock, D., McKitterick, R. and Dumville, D. 1982, Ireland in Early 
Mediaeval Europe. Studies in Memory of Kathleen Hughes, Cambridge. 

Wickham, C. 1981, Early Medieval Italy. Central Power and Local Society 
400-1000, London. 

Williams, 1. 1982, Armes Prydein: The Prophecy of Britain, Dublin, Institute 
for Advanced Studies. 

WonnakLP. 1977, 'Lex Scripta and Verbum Regis: Legislation and Germanic 
Kingship from Euric to Cnut', in Sawyer and Wood, 1977, 105-38. 

Wormald,P. 1986, 'Celtic and Anglo-SaxonKingship: some further thoughts', 
in P. Szarmach and V. Oggins (eds), Sources of Anglo-Saxon Culture, 
Kalamazoo, Studies in Medieval Culture XX, 151-83. 

Wormald, P. 1994, Engla Lond: the Making of an Allegiance', Journal of 
Historical Sociology 1 , 1-24. 

*SP Brvldu